the impact of transformational leadership style of the
TRANSCRIPT
The Impact of Transformational Leadership Style of the School Principal on
School Learning Environments and Selected Teacher Outcomes
Alan M. Barnett B.A., Dip.Ed. (Macquarie University),
M.Ed.Admin. (The University of New South Wales)
A thesis submitted to The University of Western Sydney in partial fulfilment of the requirements for the degree of Doctor of Philosophy.
June, 2005
© A.M.Barnett, 2005.
Dedication To:
Max Knight, a wonderful principal and friend,
who demonstrated the real meaning of servant leadership.
ii
Acknowledgements
I would like to acknowledge and thank the principals and teachers who participated in this study, and the access provided by the New South Wales Department of Education and Training to conduct this research in state secondary high schools.
I would also like to thank my supervisors, Professor Herbert W. Marsh and Associate Professor Rhonda Craven for their careful and thought-provoking supervision of this project through its many stages. I would also like to acknowledge and thank Dr Alan Williamson for his guidance and assistance in the area of qualitative methodology.
I would also like to acknowledge and thank my wife Kerry, and sons Stephen, Andrew and Jeremy for their continued support, practical assistance and encouragement, even when the going got tough. In the same way I thank my extended family for their affirming support.
iii Figu
Statement of Authentication The work presented in this thesis is, to the best of my knowledge and belief, original,
except as acknowledged in the text. I hereby declare that I have not submitted this
material, either in whole or in part, for a degree at this or any other institution.
………………………………………………
(Signature)
iv
CONTENTS
Page
Abstract xviii
1. Introduction 1
2. Review of the Literature: School Leadership, School Learning
Environment and Teacher Job Satisfaction
7
Introduction 7
Historical Developments in Leadership Research 7
Classical Approaches to Leadership 8
Machiavelli’s “The Prince” 8
Great Man or Trait Approaches 8
The Scientific Management Approach 9
The Human Relations Movement 10
Behavioural/Functional Approaches to Leadership 11
The Ohio State University Leadership Studies 12
Blake and Mouton’s Managerial Grid 13
Douglas McGregor’s Theory X and Theory Y 14
Situational/Contingency Approaches to Leadership 15
Fielder’s Least Preferred Co-worker
Contingency Theory
16
Other Situational/Contingency Theories 17
Vroom and Yetton’s (1973) Normative
Contingency Theory
17
Reddin’s (1970) Three Dimensional Theory 18
Hersey and Blanchard’s (1984) Situational
Theory
18
Evans’ (1970) and House’s (1971) Path-Goal
Theory
18
Summary 19
Transactional and Transformational Leadership 19
i
Leadership Research in Education 24
General Leadership Research 24
Transformational Leadership Research 26
American Community Colleges—Roueche,
Baker and Rose (1989) and Murray and Feitler
(1989)
26
School Reform and Transformational
Leadership—Leithwood and Jantzi (1990)
28
Principals’ Problem Solving Processes—
Leithwood and Steinbach (1991)
30
School Restructuring—Leithwood, Jantzi, Silins
and Dart (1992)
32
Principal Leadership in Private Schools—
Hoover, Petrosko and Schulz (1991)
34
Transformational and Transactional Leadership
style behaviours in Educational Organisations—
Kirby, King and Paradise (1992)
35
Transformational Leadership in Inclusive
Schools—Ingram (1997)
39
Principals’ Leadership Qualities—Leithwood
and Jantzi (1997)
40
Leadership and Student Engagement—
Leithwood and Jantzi (2000)
41
Transformation Leadership and Teachers’
Commitment to Extra Effort—Geijsel, Sleegers,
Leithwood and Jantzi (2002)
43
Summary 44
Leadership Constructs Used in this Thesis 44
Leadership Summary 47
School Learning Environment 48
The Historical Development of School Learning Environment
Research
48
School Effectiveness and School Learning Environment 48
ii
Measuring School Learning Environment 52
The Behavioural Approach to Describing School
Learning Environment
52
The Organisational Dynamics Approach to Describing
School Learning Environment
54
Pupil Control Approaches to School Learning
Environment: Custodial to Humanistic Controls
55
The National Association of Secondary School
Principals’ School Climate Survey
56
Profile of Organisational Characteristics (POC)
Measure
57
Summary 59
The Relationship between School Learning Environment,
School Culture and School Effectiveness
59
Summary 61
School Learning Environment Constructs used in this Thesis 61
School Learning Environment Research 64
School Level Environment Questionnaire (SLEQ)
Research
66
Summary 70
Teacher Job Satisfaction 71
The Historical Development of Teacher Job Satisfaction
Theory
71
Maslow’s (1943) Hierarchy of Needs Model 71
Herzberg, Mausner and Snyderman’s (1959) Two-
Factor Theory
73
Teacher Job Satisfaction Constructs used in this Thesis 75
Summary 77
Leadership Behaviour, School Learning Environment and
Teacher Job Satisfaction
77
Theoretical Framework 79
Summary 83
iii
3. Aims, Hypotheses, Research Questions: Rationale 84
Introduction 84
The Problem 84
Aims 85
Statement of the Hypotheses and Research Questions 86
Hypothesis 1: The Relation of Differential Leadership
Behaviours and School Learning Environment
86
Hypothesis 1.1 86
Hypothesis 1.2 86
Hypothesis 2: The Relation of Differential Leadership
Behaviours and Teachers’ Satisfaction with their
Leader, Teachers’ Self-Perceived Effectiveness,
Influence and Control
87
Hypothesis 2.1 87
Hypothesis 2.2 87
Research Question 3: What Relation do Antecedent
Background Variables have with School Leadership,
School Learning Environment and Teacher Outcomes?
87
Research Question 3 87
Research Question 4: What are the Leadership
Behaviours that Enhance or Erode Teachers’
Perceptions of the School Learning Environment,
Satisfaction, Effectiveness, Influence and Control?
87
Research Question 4.1 87
Research Question 4.2 88
Research Question 4.3 88
Rationale for Hypotheses and Research Questions 88
Rationale for Hypotheses 1.1 and 1.2: The Relation of
Differential Leadership Behaviours and Learning
Environment
88
Rationale for Hypotheses 2.1 and 2.2: The Relation of
Differential Leadership Behaviours and Teachers’
Satisfaction with their Leader, Teachers’ Self-Perceived
89
iv
Effectiveness, Influence and Control
Rationale for Research Question 3: What Relation do
Antecedent Background Variables have with School
Leadership, School Learning Environment and Teacher
Outcomes?
90
Rationale for Research Questions 4.1, 4.2 and 4.3: What
are the Leadership Behaviours that Enhance or Erode
Teachers’ Perceptions of the School Learning
Environment, Satisfaction, Effectiveness, Influence and
Control?
91
Research Question 4.1. 91
Research Question 4.2. 92
Research Question 4.3. 92
Summary 93
4. Methodology 94
Introduction 94
Study 1: Quantitative Methodology 94
Purpose 94
Sample Selection 95
Participants 96
Instrumentation 96
Multifactor Leadership Questionnaire
MLQ-5X (Short)
96
School Level Environment Questionnaire
(SLEQ)
98
Teacher Outcomes 98
Demographic Survey 100
Procedures 100
Overview of Methodology Employed to Test Research
Hypotheses Proposed
100
Hypotheses 1.1 and 1.2: The relation of
differential leadership behaviours and school
100
v
learning environment
Hypothesis 2.1 and 2.2: The relation of
differential leadership behaviours and teachers’
satisfaction with their leader, teachers’ self-
perceived effectiveness, influence and control
101
Research Question 3: What Relation do
Antecedent Background Variables have with
School Leadership, School Learning
Environment and Teacher Outcomes?
101
Data Analysis Procedures 102
Multilevel modelling 102
Levels of Analysis 107
Study 2: Qualitative Methodology 107
Purpose 107
Theoretical Background 109
Research Design 109
Developing a Conceptual Framework 110
Sample Selection 110
Instrumentation 111
Participants 112
Overview of Interviewing Approaches 112
Overview of the Methodology Employed to test
Research Questions Proposed
116
Data analysis 117
Unit of analysis 118
Case analysis versus cross-case analysis 118
Summary 119
5. Study 1 Results: The Nature and Relation of Principal Leadership
Behaviours and School Learning Environments and Teacher
Outcomes
120
Introduction 120
Development of Constructs Used in this Analysis 120
vi
Results: Hypothesis 1.1 and 1.2: The Relation of
Transformational and Transformational Leadership and School
Learning Environment
121
Overview 121
Analysis 121
Student Supportiveness 121
Affiliation 133
Professional Interest 136
Centralisation 140
Innovation 144
Resource Adequacy 147
Achievement Orientation 152
Hypothesis 1.1: Results 156
Hypothesis 1.2: Results 157
Hypothesis 1.1. and 1.2. Summary 158
Results: Hypothesis 2.1 and 2.2: The Relation of
Transformational and Transactional Leadership Behaviour and
Teachers’ Satisfaction with their Leader, Teachers’ Self-
Perceived Effectiveness, Influence and Control
160
Overview 160
Analysis 160
Global Satisfaction with Leadership 160
Teachers’ Perception of Influence 164
Teachers’ Perception of Effectiveness 169
Teachers’ Perception of Control 173
Hypothesis 2.1: Results 177
Hypothesis 2.2: Results 177
Hypothesis 2.1. and 2.2: Summary 178
Results: Research Question 3. What Relation do Antecedent
Background Variables have with School Leadership, School
Learning Environment and Teacher Outcomes?
179
Overview 179
Analysis 179
vii
The Effect of Antecedent Variables on School
Leadership
180
Vision 180
Individualised Consideration 182
Laissez-Faire Leadership 184
The Effect of Antecedent Variables on School Learning
Environment and Teacher Outcomes
186
School Level Results 186
School Leadership variables 186
School Learning Environment variables 187
Teacher Outcome variables 188
Teacher Level Results 188
School Leadership variables 188
School Learning Environment variables 189
Teacher Outcome variables 190
Research Question 3: Summary 191
School Level 191
Teacher Level 192
Summary 194
6. Study 2 Results: A Qualitative Perspective on the Effects of
Principals’ Leadership Behaviour on Aspects of School Learning
Environment and Selected Teacher Outcomes
197
Introduction 197
Procedures Undertaken 198
Sample Selection 199
School Selection 199
Background of the Selected Schools 200
School A 200
School B 200
School C 201
Informants 202
Background of Informants 202
viii
Principals 203
School A 203
School B 203
School C 203
Teacher Informants: School A 203
Respondent 1 203
Respondent 2 204
Respondent 3 204
Respondent 4 204
Teacher Informants: School B 205
Respondent 1 205
Respondent 2 205
Respondent 3 205
Teacher Informants: School C 205
Respondent 1 205
Respondent 2 206
Respondent 3 206
The Interview Instrument 206
Data Collection 207
Pre Visitation 207
Interviews 208
Interview Protocols 208
Data Analysis 209
Presentation of Analysed Data 210
Aspects of Transformational and Transactional Leadership 211
School Leadership—Vision 211
Ownership of and Commitment to the Vision 211
Elements of the Vision 215
Implementation of the Vision 218
The Centrality of the Vision 222
Research Question 4.1: Summary of Visionary
Leadership Results
225
School Leadership—Individualised Consideration 226
ix
“Do You Feel the Principal Knows You?” 226
“Do Principals know their staff?” 228
Strategies employed to know staff 230
The Principal’s Accessibility 233
Research Question 4.1: Summary of Individualised
Consideration Results
235
Teacher Job Satisfaction 236
Strategies used by Principals to Achieve Excellence 236
How Principals let Staff know that they’ve done a Good
Job
242
What Staff Believe makes Working in their School
Satisfying
245
Research Question 4.2: Summary of Teacher Job
Satisfaction Results
249
School Learning Environment 250
Important Attitudes and Values Regarding School
Learning Environment
250
Staff Attitudes to Being in their School 254
Creating and Enhancing a Positive Learning
Environment
257
School Leadership and the School Learning
Environment
261
Research Question 4.3: Summary of School Learning
Environment Results
266
Qualitative Findings on Transformational and Transactional
Leadership Styles
267
Summary 273
7. Discussion and Implications for further Research and Educational
Leadership Practice
274
Introduction 274
Discussion of Findings 275
Discussion of Aim 1: Testing Differential Leadership 275
x
Style
Vision 275
Individualised Consideration 278
Laissez-Faire Leadership 282
Summary Aim 1: Differential Leadership
Attributes
282
Discussion of Aim 2: The Differential Impact of
Leadership Styles on School Learning Environment
282
Summary Aim 2: Differential Impact on School
Learning Environment
285
Discussion of Aim 3: The Differential Impact of
Leadership Styles on Perceptions of Satisfaction,
Influence, Effectiveness and Teacher Control
286
Summary Aim 3: The Differential Impact of
Leadership Styles on Satisfaction, Influence,
Effectiveness and Teacher Control
288
Discussion of Aim 4: The Differential Impact of
Antecedent Variables on Leadership Styles and School
Learning Environment
289
School size 289
Teacher position 290
Teacher time in current school 290
Teacher gender 291
Time spent in a school with current principal 291
Principal gender 292
Summary Aim 4: The Differential Impact of
Antecedent Variables
292
Discussion of Aim 5: Identify leadership strategies that
enhance or erode teachers’ perceptions of school
learning environment and teacher satisfaction
292
Enhancing perception of school environment 292
Eroding perceptions of school learning
environment
295
xi
Summary Aim 5: Identify leadership strategies
that enhance or erode teacher’s perception of
school learning environment and teacher
satisfaction
297
Strengths and Limitations of the Study 297
Implications for Research and Theory in Educational
Leadership
299
Summary 300
8. Summary and Conclusion 302
References 306
Appendices 322
xii
Appendixes
4.1 Permission To Conduct Research
323
4.2 Instructions Regarding Involvement in this Research
326
4.3 Teacher Questionnaires
330
4.4 Principal Questionnaire
342
4.5 Interview Protocols: Teachers and Principal
348
4.6 Informed Consent
354
5.1 Development of Leadership, School Learning Environment and Teacher Outcomes Models
355
5.2 Demographic Characteristics of Participants, Study 1 396
xiii
List of Tables 4.1 Summary of Instruments—Multifactor Leadership Questionnaire—
MLQ-5X (Short), School Level Environment Questionnaire (SLEQ) and Teacher Satisfaction Questionnaire
99
5.1 Variation in Teachers’ Student Supportiveness Scores (N = 458) in 52 Schools
126
5.2 Variation in Teachers’ Affiliation Scores (N = 458) in 52 Schools
134
5.3 Variation in Teachers’ Professional Interest Scores (N = 458) in 52 Schools
138
5.4 Variation in Teachers’ Centralisation Scores (N = 458) in 52 Schools
141
5.5 Variation in Teachers’ Innovation Scores (N = 458) in 52 Schools
145
5.6 Variation in Teachers’ Resource Adequacy Scores (N = 458) in 52 Schools
148
5.7 Variation in Teachers’ Achievement Orientation Scores (N = 458) in 52 Schools
153
5.8 Variation in Teachers’ Global Satisfaction with Leadership Scores (N = 458) in 52 Schools
161
5.9 Variation in Teachers’ Perception of Influence Scores (N = 458) in 52 Schools
166
5.10 Variation in Teachers’ Perception of Effectiveness Scores (N = 458) in 52 Schools
171
5.11 Variation in Teachers’ Perception of Control Scores (N = 458) in 52 Schools
174
5.12 Variation in Teachers’ Perception of Principal’s Vision Scores (N = 458) in 52 Schools
181
5.13 Variation in Teachers’ Perception of Principal’s Individualised Consideration Scores (N = 458) in 52 Schools
183
5.14 Variation in Teachers’ Perceptions of Principal’s Laissez-Faire Leadership Scores (N = 458) in 52 Schools
185
6.1 Conceptual framework for the qualitative research
199
6.2 Comparison of surveyed schools
201
6.3 Demographic characteristics of informants 202
xiv
6.4 Ownership and Commitment to the Vision
213
6.5 Elements of the Vision
216
6.6 Implementation of the Vision
219
6.7 The Centrality of the Vision
224
6.8 “Do you feel the principal knows you?”
227
6.9 “Do you know your staff?”
229
6.10 Strategies employed by the principal to get to know the staff
231
6.11 “Do you feel the principal is approachable?”
234
6.12 Strategies used by principals to achieve excellence
237
6.13 How principals let staff know that they’ve done a good job
244
6.14 What staff believe makes working in their school satisfying
247
6.15 Important attitudes and values regarding school learning environment
252
6.16 Staff attitudes to being in their school
256
6.17 Creating and enhancing a positive learning environment
258
6.18 School leadership and the school learning environment
263
xv
List of Figures 2.1 General theoretical framework guiding research on leadership, school
learning environment and selected teacher outcomes
79
2.2 Specific conceptual framework guiding research on leadership, school learning environment and selected teacher outcomes
82
4.1 Mathematical representation of baseline variance components model
104
4.2 Mathematical representation of variance components models with explanatory variable
106
5.1 A baseline, two level variance component model for student supportiveness
122
5.2 Residuals of school level student supportiveness (SSs) scores, ranked for 52 schools, showing 95% confidence intervals
123
5.3 The effect of leadership explanatory variables on teachers’ student supportiveness scores (Multilevel Model 2)
124
5.4 The effect of school average demographic explanatory variables on teachers’ student supportiveness scores (Multilevel Model 3)
125
5.5 The cumulative effect of teacher level demographic explanatory variables on teachers’ student supportiveness scores (Multilevel Model 4)
128
5.6 The cumulative effect of school level leadership explanatory variables on teachers’ student supportiveness scores (Multilevel Model 5)
129
5.7 The cumulative effect of school level school learning environment explanatory variables on teachers’ student supportiveness scores (Multilevel Model 6)
130
5.8 Statistically significant explanatory variables that influence teachers’ student supportiveness scores (Reduced Model 6)
131
5.9 Residuals of school level affiliation (AFFs) scores, ranked for 52 schools, showing 95% confidence intervals
133
5.10 Residuals of school level professional interest (PIs) scores, ranked for 52 schools, showing 95% confidence intervals
139
5.11 Residuals of school level centralisation (CENs) scores, ranked for 52 schools, showing 95% confidence intervals
142
xvi
5.12 Residuals of school level innovation (INNs) scores, ranked for 52 schools, showing 95% confidence intervals
146
5.13 Residuals of school level resource adequacy (RAs) scores, ranked for 52 schools, showing 95% confidence intervals
149
5.14 Residuals of school level achievement orientation (AOs) scores, ranked for 52 schools, showing 95% confidence intervals
152
5.15 Residuals of school level teachers’ perception of global satisfaction with leadership (SAT) scores, ranked for 52 schools, showing 95% confidence intervals
162
5.16 Residuals of school level teachers’ perception of influence (INFLU) scores, ranked for 52 schools, showing 95% confidence intervals
165
5.17 Residuals of school level teachers’ perception of effectiveness (EFF) scores, ranked for 52 schools, showing 95% confidence intervals
170
5.18 Residuals of school level teachers’ perception of control (TCON) scores, ranked for 52 schools, showing 95% confidence intervals
175
xvii
Abstract
Much of the recent literature on effective schools has intuitively connected
the leadership role of the school principal, and school learning environment, to the
achievement of organisational outcomes such as those related to teacher
performance. Transformational leadership theory has cast these relations in a new
perspective, where advocates have claimed that transformational leaders are more
able to manipulate environmental contexts so as to achieve their organisational
objectives compared to transactional and laissez-faire leadership styles.
This study examines the effects of different types of secondary principals’
leadership behaviours on aspects of a school’s learning environment, and selected
teacher outcomes. Study 1 involved a quantitative analysis of teachers’ perceptions
of principals’ leadership style, school learning environment and selected teacher
outcomes, using an instrument administered in 52 randomly selected schools and
involving 458 teachers from across New South Wales, Australia. Study 2 involved a
qualitative analysis of data collected from 12 respondents in three schools,
examining those leadership practices that enhanced or eroded teachers’ perceptions
of school learning environment and teacher satisfaction. Specifically, the qualitative
phase of the study was used to investigate those specific principal leadership
behaviours that enhance both teacher outcomes and perceptions of school learning
environment.
A synergy was achieved by undertaking two studies drawing upon a multi-
method approach. While some transformational leadership behaviours of vision
building were demonstrated to be effective in influencing school learning
environment and teacher outcome variables, it was a combination of transformational
and transactional leadership styles that demonstrated the most impact in relation to
school learning environment and teacher outcomes.
These results are significant for those who exercise leadership authority in
schools, and are contrary to the findings suggested by transformational leadership
literature. Practitioners will welcome the opportunity to tailor leadership behaviours
xviii
to achieve specific school learning environment and teacher outcome objectives,
while those involved in principal training will recognise the potency of the
behavioural aspects of the transformational and transactional paradigms.
xix
CHAPTER 1
INTRODUCTION
The study of what makes effective schools has dominated much of the
educational research agenda over the past two decades (Caldwell, 1998; Hallinger &
Heck, 1998; Johnson, Livingston, Schwartz & Slate, 2000). While many factors have
been posited as assisting in the achievement of desired educational outcomes (Fullan
& Hargreaves, 1991; Ingram, 1997; Leithwood 1992; Wyatt, 1996), questions have
been raised as to whether this research has improved learning outcomes for students
(Wyatt, 1996).
A focus of many effective schools studies has been on the principal’s
leadership role in enhancing organisational performance (Shum & Cheng, 1997;
Starrett, 1993). Leadership can be broadly defined as the process of moving a group
or groups in some direction through mostly non-coercive means (Kotter, 1988).
Recent studies have highlighted the mediating role principals serve between teachers
and learners (Silins & Murray-Harvey, 1999). Interestingly, results from these
studies suggest that principals have the ability to affect student achievement
indirectly by improving the tone or learning environment of a school (Johnson,
Livingston, Schwartz & Slate, 2000).
While the concepts of school leadership and school learning environment are
intertwined, few studies have related both these concepts (Griffith, 1999).
Furthermore, previous studies have been restricted in their ability to differentiate the
effects of school leadership on multifaceted school learning environment constructs.
Also, they have failed to take into consideration the hierarchical nature of effective
schools data, and have therefore been of limited predictive value (Wyatt, 1996).
Researchers have identified a redefinition in the role of the principal over the
past forty years (Lockwood, 1997). This role evolution has brought with it new
complexity, particularly in reference to the principal’s relationship with the system
office. Principals are now responsible for encouraging democratic decision making,
facilitating school improvement and promoting self management, as well as for staff
1
selection and developing a market orientation (Caldwell, 1998; Elliot, 1992; Silins,
1994). Turbulence within the prevailing social, moral and economic environment
(Davies, 2002; Starrett, 2004a; Vandenberghe, 1995) has fostered a market-driven
approach to educational accountability that has emphasised the need for schools to
produce outcomes over and above what would normally be predicted, given family
background and socioeconomic status (Mann, 1992). These “within-school”
practices are considered highly desirable because they do not “depend on extra
money or new grants of authority” to achieve their “value-added” goals (Mann,
1992, p. 224). Three useful “within-school” practices identified in the effective
schools literature are germane to this present study. They include the identification of
strong leadership at the school level, best practice teaching pedagogy, and an
organisational environment conducive to fostering the work done by teachers
(Lezotte, 1991; Wyatt, 1996).
In effective schools, the principal acts as an instructional leader who
persistently communicates the organisational vision and mission of the school to the
stakeholders of the community (Lezotte, 1991). Incorporated into the leadership
function of the principal is the role of creation of shared values among the
community members. The principal defines the vision of the school, which continues
to act as “magnetic north” in setting the direction for the community to travel. To
achieve this, however, the principal recognises that, in a complex school community,
they are a “leader of leaders” rather than a “leader of followers”. The effective school
principal is also concerned to provide instructional leadership that emphasises best
practice teaching pedagogy. Lezotte (1991) defined this leadership as the allocation
of significant amounts of instructional time, and the promotion of teaching methods
favourable to the attainment of student mastery of content. Further, principals act as
instructional coaches who seek constantly to monitor value-added results in student
achievement (Wyatt, 1996).
Effective schools also have an organisational environment that is conducive
to fostering the work done by teachers. This includes the creation of an environment
which is orderly, purposeful and businesslike, and free from the threat of physical
harm (Lezotte, 1991). In this environment, there is a climate of expectation that
encourages all students to achieve high levels of mastery (Wyatt, 1996). In these
2
effective school environments, there is an attitude among teachers that intrinsic job
satisfiers such as pupil achievement, teacher achievement, pupil growth, recognition,
mastery and self-growth and positive relationships will outweigh the extrinsic
dissatisfiers, including rapid change, poor supervision and administrative workloads
(Dinham & Scott, 1998).
Current research in the effective schools area acknowledges the difficulty in
linking principals’ leadership practices directly with student achievement (Hallinger
& Heck, 1998; Lockwood, 1997). However, several studies highlight the indirect
effects a principal’s leadership practices have on teacher motivation, commitment
and innovation (Blasé & Kirby, 1992), faculty trust (Griffith, 1999) and the teaching
and learning cycle (Silins & Murray-Harvey, 1999). Hallinger and Heck (1998),
commenting on fifteen years of effective schools research, concluded that effective
principals influence student learning by manipulating the internal processes and
contextual factors of a school (Griffith, 1999; Silins, 1994). Although it is recognised
that principal leadership is an essential component in determining school learning
environment, the relationships between the leadership style of the principal, teaching,
student achievement (Blasé & Blasé, 2000), and teacher job satisfaction (Dinham &
Scott, 1998) within their school contexts have not been adequately studied (Griffith,
1999). Further, recent paradigm shifts in conceptualising leadership have also
encouraged educational researchers to consider these relationships from the
perspective of new leadership models. Prominent among them are the
transformational and transactional leadership models (Bass, 1985a; Bass & Avolio,
1997; Burns, 1978; Leithwood & Jantzi, 1990; Sergiovanni, 1990a).
Transformational leadership is hypothesised to occur when leaders and
followers unite in pursuit of higher order common goals, when “one or more persons
engage with others in such a way that leaders and followers raise one another to
higher levels of motivation and morality” (Burns, 1978, p. 20). This implies that the
leader-follower relationship is one in which the purposes of both become fused,
creating unity and collective purpose (Barker, 1990). The leader motivates followers
to “work for transcendental goals instead of immediate self-interest, for achievement
and self-actualisation rather than safety and security” (Murray & Feitler, 1989, p. 3),
and creates within followers a capacity to develop higher levels of commitment to
3
organisational goals (Leithwood & Jantzi, 2000). Transactional leadership is
hypothesised to occur when there is a simple exchange of one thing for another.
Burns (1978, p. 19) argued that transactional leadership occurs “when one person
takes the initiative in making contact with others for the purpose of exchange of
valued things”. In this relationship the leader and the led exchange needs and
services in order to accomplish independent objectives (Barker, 1990; Kirby, King &
Paradise, 1992). Bass (1985) conceptualised a third type of leadership, laissez-faire
leadership, which is hypothesised to occur when there is an absence or avoidance of
leadership. In this case decisions are delayed, and reward for involvement is absent.
No attempt is made to motivate followers, or to recognise and satisfy their needs
(Bass & Avolio, 1997).
Effective schools researchers agree that the school learning environment is
directly influenced by the leadership style of the principal (Dinham, Cairney, Craigie
& Wilson, 1995; Griffith, 1999; Leithwood, Begley & Cousins, 1990). The
leadership style of the principal is largely responsible not only for the teaching and
learning environment of a school, but also for the performance of the staff beyond
that which exceeds a basic level (Lambert, 1990). Australian research has
demonstrated that it is the style of leadership exercised by the principal which
determines not only the tone of the entire school, but also the quality of education
offered by a school (Spence, 1991). Transformational leadership models also
emphasise that “transformational leaders are able to alter their environments” to meet
their desired outcomes (Kirby, King & Paradise, 1992, p. 303). Transformational
school leaders do this by promoting educational restructuring and innovation,
focusing on building vision, encouraging collaborative participation and raising the
role of follower to that of leader (Silins, 1994).
The primary purpose of this study is, firstly, to contribute to addressing some
of these issues by capitalising upon theoretical models of transformational and
transactional leadership to elucidate the impact of secondary principals’ leadership
styles in terms of enhancing the school learning environment and teachers’ self-
perceptions of their job satisfaction (Bass, 1985a; Burns, 1978; Sergiovanni, 1991).
Few educational studies have related the concepts of school leadership and school
learning environment (Griffith, 1999), particularly from a transformational and
4
transactional leadership paradigm. The potential yield from the application of this
model is worth noting as it is suggested that the addition of transformational
leadership style behaviours in principals can produce extraordinary and increased
performance in teachers. Moreover, proponents claim that transformational
leadership style is a behavioural process which is readily learned, in which case this
study potentially may have important implications for the training and professional
development of principals (Barth, 1990; Bass, 1985a; Burns, 1978; Conger &
Kanungo, 1988; Sergiovanni, 1990a). Any study that is aimed at applying and
improving styles of leadership provided by school principals has to be judged as
worthwhile. Many principals are simply not trained to deal with the problems facing
today’s schools, and they seldom have time to reflect on their practices and bring a
critical perspective to their work (Starrett, 2004b).
Secondly, this study examines the transformational and transactional
leadership paradigm within an Australian school context (Parry & Sarros, 1996).
While several overseas studies have focused on the transformational and
transactional leadership paradigm in organisational settings as diverse as accounting
firms, bank and financial offices, private colleges, schools and universities, and
within the military (Avolio, Bass & Jung, 1995), few studies have been undertaken
generally in New South Wales and specifically within secondary schools.
More specifically the study aims to:
a) test whether the leadership behaviours as predicted by Bass et al. (1997) can be
identified in a sample of New South Wales (NSW) secondary schools;
b) test the differential impact of transformative, transactional and laissez-faire
leadership styles of secondary principals on school learning environment;
c) test the differential impact of transformative, transactional and laissez-faire
leadership styles of secondary principals on teachers’ self-perceptions of
satisfaction with their principal, and teachers’ self-perceived effectiveness to
produce desirable educational outcomes;
d) examine the relation between different types of leadership style (transformative,
transactional and laissez-faire) of secondary principals, school learning
environment constructs and antecedent variables;
5
e) identify, using qualitative data gathered from respondents, leadership strategies
that enhance and/or erode teachers’ perceptions of school learning environment,
and teacher job satisfaction.
These aims are achieved by employing a synergistic blend of quantitative and
qualitative research methods and the strongest available statistical and qualitative
data analysis tools, in the context of a sound research design based on an adequate
sample size. This research will involve two studies (one qualitative and the other
quantitative) in order to examine the questions raises.
6
CHAPTER 2
REVIEW OF THE LITERATURE:
SCHOOL LEADERSHIP, SCHOOL LEARNING
ENVIRONMENT AND TEACHER JOB SATISFACTION
Introduction
The purpose of this chapter is to review literature pertinent to the present
investigation in order to place the focus areas of this study—school leadership,
school learning environment and teacher job satisfaction—into their historical and
theoretical context, and to trace relations between these areas. Each of these key
focus areas is discussed in turn. Firstly, historical developments in the focus area are
discussed. Secondly, an overview of the nature of the focus area is provided. Thirdly,
recent research—both in educational and in non-educational settings—that has
influenced this current study is examined, together with the implications of this
research for the current study. Finally, consideration is given to the focus area
constructs presented in this study.
Historical Developments in Leadership Research
Leadership has long been a subject of consideration for both researchers and
philosophers alike, and to many, the study of history is simply the study of leaders
(Bass, 1990). Ancient Egyptian, Greek, Roman and Chinese cultures all devoted
much thought to the principles of leadership, in an effort to find generalisations that
could be applied to current leaders from past leaders. Philosophers developed notions
of leadership they felt were essential to a thorough understanding of the subject—
Ptahhotep (ancient Egypt): authority, perception and justice; Confucius (ancient
China): moral example and manipulative reward; and Homer (ancient Greece):
judgment, justice, wisdom, counsel and valour (Yukl, 2002).
7
Classical Approaches to Leadership
Machiavelli’s “The Prince”. Machiavelli’s (1513) “The Prince” stands as a
work of insight into the functioning of the esecuzione or executive (English, 1992).
Machiavelli pragmatically acknowledged that a leader must succeed at all cost. If a
prudent “prince” cannot rule, then he “should not honour his word when it places
him at a disadvantage” (Kellerman, 1984, p. 84). Machiavelli argued that most
people resist law and reason, and consider the exercise of power to be unjust, no
matter who wields it. Therefore, power must be rendered in someone else’s name.
So, for the king, rule was by “divine right”, or by natural right, through the law of
nature. Machiavelli’s answer to the dilemma of compliance and authority was in the
creation of an executive, where a “person rules for something or someone else and
thus can disclaim having any true power at all except that which is bestowed by third
parties” (English, 1992, p. 12). Machiavelli argued a strong executive needed
steadiness, firmness and a concern for the maintenance of authority, power and
government, and it was best if the objectives of leadership could be achieved by
winning the esteem of the people. If they could not, craft, violence, threat and
treachery were justified (Bass, 1990). In schools, the principal can be seen as the
embodiment of Machiavelli’s “executive”. The principal was empowered to “rule” a
school on behalf of a third party (English, 1992). In Australia, this took the form of a
State or Federal Government Department of Education, which in turn acted on behalf
of the Crown. Machiavelli’s model, which emphasised the devolution of authority
and leadership, served most organisations, including education organisations, until
the early 1920s, when new leadership theories emerged.
Great man or trait approaches. Leadership studies at the turn of the 20th
century focused on the perceived possession of certain traits that led to the success of
leaders (Adair, 1984). Aristotle believed “from the moment of birth, some are
marked out for subjection, others for rule” (Hoy & Miskel, 1996, p. 376). This
formed the basis for the “great man” or “trait” theories of leadership. A “trait” was
defined as “any distinctive physical or psychological characteristic of the individual
to which the individual’s behaviour can be attributed” (House & Baetz, 1990, p. 8). It
was thought that although these characteristics might well be developed to some
8
extent, they “were basically something you either had or did not have” (Beckner,
1990, p. 8). Leaders, it was said, were born, not made (Northouse, 2004).
Early research centred on identifying leadership traits, but this methodology
suffered from two major drawbacks. Firstly, despite several experimental studies,
there was no agreement regarding the qualities that allowed a leader to have
dominance over his or her followers. Secondly, the trait or qualities approach to
leadership was not suited to leadership training as intrinsically it favoured the
selection of leaders, as opposed to the training of leaders (Bass, 1990).
After reviewing 124 different leadership studies, the trait approach was all
but shelved after researcher Ralph Stogdill (1948) argued that this approach yielded
negligible and confusing results (Hoy & Miskel, 1996). Several of the traits that were
found to be crucial in some studies were found to be unimportant in other studies.
While Stogdill (1948) concluded that there is a cluster of personality traits that can
differentiate leaders from non-leaders, and effective leaders from non-effective
leaders (House & Baetz, 1990), his review failed to find support for the notion that
leaders must possess a particular set of traits to be successful in leadership. Stogdill
(1948) argued that the importance of individual traits was dependent on particular
situations, and that no traits could be identified as being necessary for ensuring
successful leadership (Yukl, 2002). Subsequent reviews supported Stodgill’s (1948)
conclusion that the use of personality or trait characteristics to differentiate
successful leaders was remarkably unsuccessful (Hoy & Miskel, 1996; Beare,
Caldwell & Millikan, 1989; Kellerman, 1984).
The scientific management approach. Frederick Taylor (1856-1915)
developed an approach to leadership based on the ideals of increased output by
means of technological management and efficiency. Central to Taylor’s scientific
management philosophy was the concept of the machine metaphor (Hoy & Miskel,
1996, p. 9). Efficiency was the supreme goal, and workers were expected to “adjust
to rationally planned and executed management” (Beckner, 1990, p. 8). The leader’s
task was to pursue the needs of the organisation rather than the needs of the workers,
using rationally planned flow charts and job descriptions.
9
Taylor (1911) argued that organisations, including schools, could be seen as
“closed-systems”, where the forces affecting decision-making were fixed and
predictable, and that organisations could operate in an environment of certainty
(Owens, 1998, p. 42; Short & Greer, 2002, p. 2). Such organisations could, by virtue
of closed system rationality, operate in environments of certainty that would permit
the identification of operational objectives, and develop plans, strategies and
mechanisms to meet and maximise those objectives. Decision-making in “open-
systems” was argued to be “hopelessly irrational and organisational environments
hopelessly turbulent and ambiguous” (Sergiovanni & Carver, 1980, p. 59).
In educational settings, the role of the principal as leader under a scientific
management approach was to define the goals of the school, and coordinate the
labour of teachers to meet those goals. This requires the principal finding the best
method of work, determining qualification standards necessary to complete the work,
keeping teachers supplied with necessary materials to finish tasks, and providing the
necessary incentives to stimulate desirable efforts (Short & Greer, 2002). Under
Taylor’s (1911) philosophy, the school became a factory, and the teacher, a worker
on the assembly line, whose function was to perform some process on the student,
who was the product in this context (Owens, 1998).
The human relations movement. By the late 1920s and early 1930s,
researcher Elton Mayo argued that finding the best technological methods in the
productive processes was not enough. Attention needed to be given towards human
relations. These included cooperative goal attainment among workers, the
consideration of human perspectives and feelings, and provision of opportunities for
personal growth. The Human Relations approach emphasised that “concern for task
(output) should be replaced with emphasis on concern for relationships (people)”
(English, 1992, p. 8).
The development of the Human Relations approach is accredited to a series
of experiments undertaken in the Hawthorne plant of the Western Electric Company
of Chicago. Productivity was measured in groups of workers who worked under
differing lighting conditions. The results were puzzling, because increases in
production rates did not coincide with increases in illumination intensity. The
10
productivity of a test group whose illumination intensities varied was compared to
the productivity of a control group whose illumination intensities remained constant.
Both groups showed productivity increases that were almost identical. Different
intensities of lighting were experimented with, and productivity measured between
the two groups. The efficiency of both groups increased, and it was noted that “the
production rates increased in the test group until the light became so poor that the
workers complained they could no longer see what they were doing” (Hoy & Miskel,
1996, p. 12).
Harvard professors Elton Mayo and Fritz Roethlisberger were consulted, as
the company suspected that both psychological and physiological factors were
involved. Through a series of further research experiments, one generalisation
became evident. Workers’ behaviour did not conform to job specifications outlined
by the company. “The group had developed an informal social structure with norms,
values, and sentiments that affected performance” (Hoy & Miskel, 1996, p. 14).
Workers responded to management as members of informal groups, not as
individuals. Further, workers used informal organisations or networks to protect
themselves against arbitrary management decisions. Lastly, Mayo and Roethlisberger
found that informal leaders were often as important as formal supervisors (Hoy &
Miskel, 1996).
The Human Relations approach challenged the validity of Taylor’s Scientific
Management approach. Furthermore, the dichotomy between task and relations
became the basis for further research on leadership (English, 1992). This research
emphasised the necessity of better understanding the nature of human needs at work,
and of applying this knowledge to the development of more effective administrators,
managers and supervisors (Owens, 1998). Important steps forward, especially in the
field of motivation, aided researchers in more fully understanding the functions of
leadership in subsequent studies.
Behavioural/functional approaches to leadership
The next line of leadership research was based on observations of leader
behaviour (Short & Greer, 2002). The failure of the trait approach to adequately
11
explain leadership, and a growing emphasis on behaviourism in psychology moved
researchers in a new direction of inquiry (Kellerman, 1984). Researchers attempted
to bring scientific methods of observation, hypothesis testing and verification to
leadership studies by analysing the behaviour of small groups both in the laboratory
and “in the field”. Many new leadership theories emerged, many of them based on
motivational theories that moved the focus of investigation away from “internal
states” (that is, leader values or personalities) to more basic questions associated with
what leaders actually do (Kellerman, 1984).
The Ohio State University leadership studies. By the 1950s, the Ohio State
University’s Personnel Research Board had developed a two-dimensional conceptual
framework for systematically describing leadership behaviour patterns (Hoy &
Miskel 1996; Northouse, 2004; Yukl, 2002). Researchers found, using extensive
empirical observations of both military and industrial leaders, that there were two
major categories of actions by which leaders’ behaviour can be described. The first
related to items of interpersonal warmth, concern for followers’ feelings and the use
of participative communications which researchers labelled as person-oriented or
“consideration” behaviour. The second category related to directiveness, goal
facilitation and task-oriented feedback, and was labelled system-oriented or
“initiating of structure” (Yukl, 2002, p. 50). Researchers imagined that these two
independent dimensions intersected at their midpoints. A leader who was system
oriented would exhibit behaviours directed primarily at the tasks of the organisation,
while a person oriented leader would exhibit behaviours primarily aimed towards
satisfying the needs and preferences of the individuals within the organisation.
Leadership behaviour could move independently along both continuums, so it was
possible to describe five identifiable regions, each with unique behavioural
characteristics. A person’s leadership style could be defined in terms of where an
individual operated on the continuums, including at the area of the intersection,
where researchers referred to a balanced or “transactional” style (Silver, 1983).
Silver (1983) highlighted two important conclusions that developed from this initial
research. Firstly, statistical treatment of early data obtained from measuring
instruments showed that the two dimensions, system and people orientated
behaviour, were significantly related, sometimes positively and sometimes
negatively. This led researchers to question the independence of the factors and the
12
nature of the measuring instrument. The second issue dealt with the development of
underlying explanatory theory. Several researchers (House, 1971; Kerr, Schriesheim,
Murphy & Stogdill, 1974; Yukl, 1971) offered theories to explain and predict
relations between leader behaviour and other phenomena, including organisation
performance.
Blake and Mouton’s Managerial Grid. Research in leadership behaviour
generated several programs aimed at management training, including Blake and
Mouton’s (1964) “Managerial Grid”. The grid was an extension of the Ohio State
University work, and was based on a 9 x 9 matrix derived from the leadership
behaviours of relationship orientation and task orientation (Hoy & Miskel, 1987).
Five leadership patterns or styles were identified (Blake & McCanse, 1991; Hoy &
Miskel, 1996; Northouse, 2004; Owens, 1998), including:
1. Relationship-oriented leadership, also called “Country Club Management”,
where production is seen as incidental to good human relations;
2. Task-oriented leadership also called “Authority-Compliance” leadership,
where good relations are incidental to high production;
3. Integrated leadership, also known as “Team Management”, where production
is achieved by integrating both task and relational aspects of production;
4. Impoverished leadership, where minimum influence is exerted in interaction
with others, and the minimum effort to get required work done is appropriate
to sustain organisational membership; and
5. Balanced leadership, also called “Middle of the Road Management”, where
the aim is to balance high productivity with good human relations.
The Managerial Grid approach served to provide a greater theoretical and
research basis for the Ohio State University’s original study. It also provided a basis
for understanding why different forms of leadership style were more appropriate in
different situations. Military leadership, for example, has a high concern for mission
and group morale, while industrial leaders are more concerned with high technical
output.
13
While the Managerial Grid approach has not been used extensively in school-
based research (Hoy & Miskel, 1996), it offers a conceptual framework that provides
a heuristic methodology for considering school leadership. This conceptualisation
provides a reasonable explanation as to why leadership style over a range of
educational levels (for example, primary and high school levels) can be different, yet
equally effective.
Douglas McGregor’s Theory X and Theory Y. McGregor’s Theory
X/Theory Y developed as a response to the scientific management belief that
supervision is the only means of assuring predictable performance (Owens, 1998).
Theory X rests on four assumptions held by an administrator or leader, which tacitly
or explicitly influence their actions in dealing with workers or followers. These are
that:
1. The average person dislikes work and will avoid it where possible;
2. Since people dislike work, they must be closely supervised (or even coerced
by the threat of punishment) in order to ensure adequate effort;
3. The average worker will shirk responsibility and will seek formal direction
from the leader; and
4. The average worker values job security above other job related factors and
has little ambition.
The alternative theory, Theory Y, is based on four assumptions that have their
basis in Herzberg’s motivational theory (Hoy & Miskel, 1996). Theory Y states that
workers have a need to self-actualise, that is, they want to obtain intrinsic, or
motivator factors, such as supervisor trust, from their jobs (Short & Greer, 2002).
The underlying assumptions are that:
1. Work is satisfying to the average person, and they will accept it as naturally
as play (recreation);
2. People will exercise initiative and self control while working if they are
committed to the objectives of the organisation;
3. The average person will, under appropriate conditions, not only accept
responsibility but will actively seek it; and
14
4. The average worker values creativity (the ability to make decisions) and will
seek opportunities to be creative.
McGregor offered Theory X/Theory Y as an explanation of real world
conditions (Owens, 1998), and focused on the importance of leader attitudes in
managing workers. The theory’s general application to administration is clear—if a
leader expects better performance, then to a limited extent, better performance will
follow.
As far as school administrators are concerned, Theory X leaders will use the
“hard sell” of authoritarian and coercive leadership style, or the “soft sell” of
paternalistic administrative behaviour to achieve their goals. Theory Y leaders will
arrange school conditions and methods of operation to support both teacher and
student effort (Hoy & Miskel, 1996). Theory X principals emphasise sign-in sheets,
checklist evaluations and strict observance of “set-in-concrete” policy. Theory Y
principals recognise the increased professionalisation of teachers and actively seek
teacher empowerment in administrative decision-making.
Situational/contingency approaches to leadership
The inadequacies of the trait approach to leadership fostered the development
of an alternative line of inquiry (Hoy & Miskel, 1996). By the late 1940s, situational
elements became the major focus for investigation. Stodgill (1948) concluded “the
evidence suggests that leadership is a relationship that exists between persons in a
social situation, and that persons who are leaders in one situation may not necessarily
be leaders in other situations” (Adair, 1984, p. 7). Hollander (1978) characterised
this approach to leadership by arguing that different situations called for different
leadership functions to be performed. Emphasis was placed on the performance of
the group’s task, and this approach highlighted the leader’s qualities that were
appropriate to the group in a given situation.
Contingency models of leadership developed as an extension to the
situational approaches and emphasised that certain leader qualities made leadership
more effective (Hollander, 1978; Hoy & Miskel, 1996; Northouse, 2004; Owens,
15
1998; Short & Greer, 2002; Silver, 1983). According to a contingency approach, it is
necessary to specify the conditions that moderate the relations between leader traits
and performance criteria. Contingency approaches emphasised relationships or social
transactions that took place between leaders and followers. Several notable
applications of this theory developed, including Fiedler’s Least Preferred Co-worker
model, Vroom and Yetton’s Normative Contingency Theory, Reddin’s Three
Dimensional Theory, Hersey and Blanchard’s Situational Theory, and Evans and
House’s Path-Goal Theory.
Fielder’s Least Preferred Co-worker Contingency Theory. Fielder’s Least
Preferred Co-worker model developed in response to the lack of empirically
demonstrated linkages between leadership style and effectiveness indicators such as
achievement (Hoy & Miskel, 1996). Fielder’s model was described as a contingency
approach, because the theory suggests that leadership effectiveness is dependent on
the fit between personality characteristics and behaviour of the leader and situational
variables such as task structure, position power and attitudes (Kellerman, 1984; Van
Fleet & Yukl, 1989). Fielder’s model is based on three postulates. Firstly, leadership
style is determined by the motivational system of the leader. Secondly, situational
control is determined by group dynamics, task structure and position power. Thirdly,
group effectiveness is a joint function of leadership style and the situation’s
favourableness, that is, group performance is contingent on the leader’s motivations
and upon the leader’s control and influence in the situation (Hoy & Miskel, 1996).
Fielder’s contingency model is built around leader style, and it distinguishes
between task-oriented and relationship-oriented leaders (Northouse, 2004). The
orientation of leaders is measured by a “Least Preferred Co-worker” (LPC) scale,
obtained by asking subordinates to “think of a person with whom they were least
able to work well” (Hollander, 1978, p. 34). This person is then rated on a set of
scales (for example, friendly to unfriendly, cooperative to uncooperative). The Least
Preferred Co-worker (LPC) score is the sum of these ratings; those low on LPC
scores are defined as being primarily task oriented, while those high on LPC scores
are considered to be primarily relationship-oriented (Kellerman, 1984). As far as
leadership training is concerned, Owens (1987, p. 142) noted that:
16
the goal of training would not be to change the basic LPC orientation of individuals but, rather, to provide them with the skills that enable them to increase the favourableness of situations in terms of their leadership style.
Therefore, development and training programs should examine ways that leaders can
better get along with subordinates, ways of handling administrative routines more
effectively, increasing technical knowledge for making decisions and ways of
assessing the favourableness of situations.
Silver (1983) noted that research regarding contingency approaches to
leadership in school settings in the published literature addressed only specific
aspects of the theory rather than the theory as a whole. Three studies of the
application of Fielder’s Least Preferred Co-worker indicated that principal’s scores
on the Least Preferred Co-worker (LPC) scales were unrelated to their self reported
behaviour patterns and willingness to take risks, a relationship the theory would
emphasise (Silver, 1983).
Hoy and Miskel (1996) reported that when Fielder’s contingency model was
applied to educational settings, the criterion of principal effectiveness had proven to
be a difficult measure to make. This was also supported by Owens (1987, p. 143),
who stated “efforts to test the applicability of Fiedler’s model to public schools tend
to encounter the traditional difficulty of assessing the effectiveness of school”.
However, Hoy and Miskel (1996) further reported that, although not statistically
significant, three studies based on Fielder’s model supported the notion that a
principal’s leadership style of relationship-orientation is associated with school
effectiveness. This finding proved consistent with general contingency theory.
Other situational/contingency theories
Vroom and Yetton’s (1973) normative contingency theory. Vroom and
Yetton (1973) developed a model for decision-making that specified the decision-
making procedures that would be most effective in each of several specific situations
(Yukl, 2002). The theoretical underpinning of this model was based on the idea that
the decision-making procedure employed by a leader would affect the quality of any
17
decisions made, and the eventual acceptance of those decisions by subordinates
expected to implement them. Vroom and Yetton reasoned that the quality of
decisions and their subsequent acceptance was dependent on various situational
aspects, and therefore a procedure that is proven effective in one situation may be
ineffective in another.
Reddin’s (1970) three-dimensional leadership theory. William Reddin
extended Blake and Mouton’s (1964) two-dimensional task and people orientation
leadership grid by adding the dimension of leadership effectiveness, in an attempt to
integrate the concepts of leadership style with specific situations. Reddin postulated
that any one of four leadership styles might be either effective or ineffective,
depending on a particular situation. While Reddin’s model extended the theoretical
work done previously on leadership traits, behaviours and groups within situational
contexts, it has not been used extensively in research. Rather, it has been used in the
training and development of administrators in various organisational environments
(Lunenburg & Ornstein, 2000).
Hersey and Blanchard’s (1984) situational leadership theory. Hersey and
Blanchard’s work (1984) further extended the original research undertaken in the
Ohio State leadership studies by theorising the relationship between leadership, task
and relationship behaviours. They included consideration of the maturity of followers
as an important situational variable in their model, arguing that the maturity of
followers will be task specific. This in turn means that differing leadership styles,
which were defined as combinations of task and relationship behaviours, would be
required if group outcomes are to be realised (Lunenburg & Ornstein, 2000; Robbins,
Millett, Cacioppe & Waters-Marsh, 1998).
Evans (1970) and House’s (1971) path-goal theory. The Path-Goal theory of
leadership used a motivational theory known as “expectancy theory” to explain how
leadership behaviours could influence the satisfaction and performance of
subordinates (Northouse, 2004, p. 123; Yukl, 2002, p. 212). Expectancy theory was
based on an understanding of worker motivation to complete assigned tasks. House
(1971) argued that subordinates made rational decisions concerning their work
assignments that ranged somewhere between maximum and minimal effort, given the
18
subordinate’s expectation of the amount of effort needed to complete the task and
achieve a desirable outcome. By considering the effects of a range of leadership style
behaviours in various situations on subordinates’ satisfaction and performance levels,
path-goal theory suggested appropriate leadership behaviours that could result in
increased subordinate motivation (Yukl, 2002).
Summary
The leadership theories considered so far have focused on the perceived
tension between factors such as autocratic as opposed to democratic leadership, or
task as opposed to relationship orientation, or directive as opposed to participative
approaches to decision making (Bass et al., 1997). While they have provided a
framework around which to consider some of the individual elements of leadership,
some researchers suggest they have failed to take into account the “strong forces”
(Bass et al., 1997) of leadership that occur when, for example, a leader shares a
strategic vision with a follower that in turn causes a change in commitment to work
(Northouse, 2004). A recent leadership paradigm, transformational and transactional
leadership, endeavours to account for these strong forces of leadership that have been
apparent in the leadership of some major educational and non-education
organisations.
Transactional and Transformational Leadership
A common element in each of the historical approaches to leadership
described earlier has been the “transactional” relationship that has existed between
leader and followers (Hoover, Petrosko & Schulz, 1991; Robbins, Millett, Cacioppe
& Waters-March, 1998). Research has confirmed that leaders have a unique,
dynamic (Bass, 1990; Kellerman, 1984) and dyadic (Lunenburg & Ornstein, 2000)
relationship with subordinates, in that “subordinates voluntarily grant power and
authority to the leader” (Owens, 1987, p. 153) and can withdraw that support from
leaders at any time. The term “transactional leadership” was used to emphasise the
two-way nature of leadership, in that leaders provide benefits for followers and in
return receive benefits from followers in a social exchange called a “transaction”
19
(Robbins et al., 1998, p. 413; Lunenburg & Ornstein, 2000, p. 150). This approach is
referred to as social exchange because both the leader and follower give something
and get something in return (Hollander, 1978). Both parties in the leader-follower
transaction agree to what must be done in order to receive reward or avoid
punishment.
James McGregor Burns (1978) refined the notion of transactional leadership
and added to it the concept of transformational leadership. Burns (1978) defined
leadership in the following terms:
Leadership over human beings is exercised when persons with certain motives and purposes mobilise, in competition or conflict with others, institutional, political, psychological, and other resources so as to arouse, engage, and satisfy the motives of followers (Burns, 1978, p. 18).
Burns emphasised “the symbiotic relationship between leadership and followership”
(Owens, 1987, p. 157), and considered leadership to be either transactional or
transformational:
The relations of most leaders and followers are transactional—leaders approach followers with an eye to exchanging one thing for another: jobs for votes, or subsidies for campaign contributions. Such transactions comprise the bulk of the relationships among leaders and followers, especially in groups, legislatures, and parties. Transforming leadership, while more complex, is more potent. The transforming leader recognises and exploits an existing need or demand of a potential follower. But, beyond that, the transforming leader looks for potential motives in followers, seeks to satisfy higher needs, and engages the full person of the follower. The result of transforming leadership is a relationship of mutual stimulation and elevation that converts followers into leaders and leaders into moral agents. (Burns, 1978, p. 4).
Transactional leadership occurs when there is a simple exchange of one thing for
another. Burns (1978, p.19) argued that it occurred “when one person takes the
initiative in making contact with others for the purpose of exchange of valued
things”. The leader and the led, exchange needs and services in order to accomplish
independent objectives (Barker, 1990). It is assumed that the leader and the follower
do not share a “common stake” in the organisation, so the wants and needs of the
follower are traded against the wants and needs of the leader, and a “bargain is
struck” (Sergiovanni, 1990a, p. 23).
20
In transformational leadership, leaders and followers are united in pursuit of
higher order common goals, and this occurs when “one or more persons engage with
others in such a way that leaders and followers raise one another to higher levels of
motivation and morality” (Burns, 1978, p. 20). This implies that the leader-follower
relationship is one in which the purposes of both become fused, creating unity and
collective purpose (Barker, 1990). The leader motivates followers to “work for
transcendental goals instead of immediate self-interest, for achievement and self-
actualisation rather than safety and security” (Lunenburg & Ornstein, 2000, p. 150;
Murray & Feitler, 1989, p. 3). Barker (1990, p. 42) argued that the only measure of
the success of the transformational leader is in the realisation of the intended social
change which satisfies “the needs and motives of both the leader and the follower”.
Further, this change must impact positively on the well being of the leader and the
follower, consistent with their values.
There are four key components central to Burns’ (1978) notion of
transformational leadership. Firstly, Burns’ (1978) used the word “engaged” in his
general definition of leader-follower relationships, as it denoted a binding,
interlocking, meshing and involvement between them. The leader-follower
relationship is one based on trust, where the leader has both a sense of belonging to
the group, and a sense of self. This “engagement” of the leader and the followers
occurs when the leader “recognises, appeals to, and acts on both her own and the
followers’ values and motivations” (Barker, 1990). The result is an arousal and
satisfaction of the motives, values and aspirations of both leader and led.
The second component of Burns’ (1978) definition of transformational
leadership involved an understanding of values, motives and needs. Burns (1978)
defined values as those enduring beliefs that induce a specific mode of conduct. They
have internalised cognitive, affective and behavioural components and functions to:
• set standards for behaviour by defining specific criteria for action;
• develop and maintain followers’ attitudes about particular things or
situations;
• justify one’s own actions and attitudes as well as allowing the judgment of
others;
21
• assist in decision making by defining specific criteria by which to make
choices between alternatives; and
• motivate people by internalising the values so deeply that they help define
personality and behaviour.
Burns (1978) based much of his theoretical work on Maslow’s hierarchy of
needs. According to Maslow’s human motivation theory (Hoover, Petrosko &
Schulz, 1991), individuals have a hierarchy of needs, beginning, in ascending order
with physiological needs, safety needs, love needs, esteem needs and finally, self-
actualisation needs. The transformational leader is able to motivate followers to go
beyond normal organisational expectations of performance (Bass, 1985a; Bass &
Hater, 1988; Harris, 1989; Hoover, Petrosko & Schulz, 1991; Lunenburg & Ornstein
2000; Sergiovanni, 1990a). This is achieved by raising followers’ consciousness
concerning the value of outcomes, encouraging transcendence of followers’ self-
interest for the sake of the organisation, and altering the need levels of the group
(Bass, 1985b; Hoover, Petrosko & Schulz, 1991).
The third component in Burns’ (1978) conceptualisation of transformational
leadership involved the tension between conflict and competition. The leader’s role
is to express, shape and capitalise on conflict instead of avoiding it. According to
Burns (1978), the way the leader uses conflict will depend on the source of that
conflict and the available resources to deal with it. Some of the skill strategies used
to shape conflict can include:
• defining the grievance and wants of followers;
• acting on behalf of followers dealing with other clusters either inside or
outside the organisation;
• influencing the intensity of the conflict;
• deviating and innovating to mediate the claims of followers;
• modifying leadership in recognition of preferences of followers;
• acting directly on behalf of followers;
• bargaining with others;
• overriding some motives of followers; and
• softening or sharpening the claims and demands of followers.
22
The last component in Burn’s (1978) conceptualisation of transformational
leadership involved understanding power. Burns (1978, p. 13) defined power as the
process:
in which power holders, possessing certain motives and goals, have the capacity to secure changes in the behaviour of a respondent, human or animal, and in the environment, by utilising resources in their power base, including factors of skill, relative to the targets of their power-wielding and necessary to secure such changes.
In terms of transformational leadership, Burns (1978) viewed both power and
leadership as relationships. He defined leadership as being a form of power, and
while raw power has the capacity to be negative, leadership is seen as positive. The
leader can secure changes in follower behaviour and in environment by utilising a
power base, which may include economic, political, institutional and personal skill.
Burns (1978) implied that power is relational among people, and therefore is
collective by nature. It involves the intention or purpose of both the leader (the
person with the power) and the recipients of the power (Barker, 1990). Burns (1978)
viewed power as a means to get things done rather than simply an end in itself. Burns
(1978) described power as having three elements: the motives and resources of the
power holder, the motives and resources of the power recipient, and the relationship
between these two. Where there are no resources to support the motives of the
leader, the leader becomes powerless. Likewise, where there is no motive to achieve
a particular end, resources diminish. Where there are no resources, motives become
dormant. Without either motives or resources, there is no power, and because both
resources and motives are in short supply, power can be viewed as elusive and
limited (Barker, 1990).
Burns (1978) viewed leadership as a form of power having a collective
purpose. If the motives of the leader do not coincide with the wants of the followers
then the leader’s power takes the form of coercion and control. This is not
transformational leadership as, central to the power of the transformational leader is
purpose, which is always in the interests of both leader and follower. Burns (1978)
conceived that leaders were either transactional or transformational. This paradigm
was modified by Bass (1985a) who argued that transformational leadership augments
the effect of transactional leadership on the efforts, performance and satisfaction of
23
followers (Bass, 1990; Sergiovanni, 1990a). Bass (1985a) suggested that many great
world leaders (for example, Kennedy and Gandhi) have moved freely between the
two extremes of transactional and transformational leadership, depending on the
dynamics of the situations involved.
The emerging perspective of transformational leadership and its associated
emphasis on cultural leadership had at its focus the development of a strong
organisational culture, in which followers believe strongly and identify personally,
and to which they gladly offer their loyalty (Owens, 1998). Transformational leaders
have the ability to unite followers in a common cause, which can produce
extraordinary performance on the part of workers (Bass, 1985a; Bass, 1985b; Beare,
Caldwell & Millikan, 1989; Harris, 1989; Hoover, Petrosko & Schulz, 1991;
Sergiovanni, 1990a; Sergiovanni, 1990b;). As far as leadership training is concerned,
transformational leadership is a completely behavioural process which is therefore
capable of being taught and learned (Bass, 1990).
The potential of transformational leadership in educational settings is
important, as schools can be regarded as cultural communities where leadership can
be considered as a shared experience or communal act (Duignan, 1989). Sergiovanni
(1990a) argued that each stage of the continuum between transactional and
transformational leadership represents a distinct school improvement phase, and
therefore links leadership style with the process of effective schooling. Further, he
says, a school principal should tactically use the transactional/transformational
continuum to ensure that different leadership styles can be used simultaneously for
different purposes or with different workers, depending on the nature of the situation.
Leadership Research in Education
General Leadership Research
Effective schools research has “constantly ascribed importance to what
principals do” (Leithwood, Begley & Cousins, 1990, p. 5). Most calls for school
reform simultaneously call for reforms in the way the principalship is conceptualised.
24
Leithwood, Begley and Cousins (1990) provided a useful overview of the empirical
research, emphasising the nexus between theory and practice, and highlighting those
aspects of principalship research that are poorly understood and would therefore
provide a fruitful focus for future studies. Leithwood et al. (1990) reviewed those
studies that dealt with the empirical evidence of the principal’s impact on teachers
and students. Among the impacts examined were positive student attitudes towards
school, achievement in basic literacy/numeracy tests and reductions in school
vandalism and absenteeism. Some impacts among teachers included use of
innovative classroom practices and teacher’s perceptions of principal leadership.
Blasé, Dedrick and Strathe (1986) examined the stress caused by the principal
and the principal’s impact on teacher classroom performance to explore correlations
with teachers’ job satisfaction. A moderately strong correlation was found between
teachers’ job satisfaction and the degree of principal’s initiation of structure (the
degree to which a leader initiates, organises and defines work related tasks).
Consideration behaviours (behaviours related to enhancing staff self-esteem) were
perceived to assist in teacher performance. Other findings highlighted
“consideration” or principal supportiveness as “the most significant predictor of staff
perceptions concerning the prevalence of group, as opposed to individual, decision
making in the school” (Leithwood, Begley & Cousins, 1990, p. 9). This
supportiveness was further found to be positively related, in teachers’ perceptions, to
principals’ involvement in curriculum decisions, the use of interactive curriculum
planning models, intimacy among staff and satisfaction with the school curriculum.
Leithwood et al. (1990, p. 9) concluded “this review suggests significant
limitations on current knowledge about the nature of principals’ impact”. However, it
could be reasonably concluded that principals have the capacity to influence both the
basic skills achievement of students, and teachers’ job satisfaction. Leithwood et al.
(1990) noted that evidence concerning other types of impacts is extremely thin. They
also suggested that subsequent research should emphasise the nature of the
principal’s impact on teachers. They noted the lack of a clear, comprehensive theory
and that future research should take the direction of a theory linking teacher
outcomes with teacher growth in classroom effectiveness. A further area of needed
research lay in the area of principal practices. Leithwood et al. (1990) suggested
25
three broad areas within this field. Firstly, research designed to test the differences
between major leadership styles and student and teacher outcomes would be
appropriate. Secondly, inquiry could be directed into the actual practices associated
with the “consideration” and “initiating structures” of leadership. Lastly, further
explorations are needed regarding the nature of effective practice in school contexts
other than turbulent urban environments already highlighted by existing literature.
Much research in educational leadership throughout the 1990s was motivated
by the expectation of an increase in understanding and knowledge using a
transformational and transactional leadership paradigm within a schools context.
Transformational Leadership Research
The theory of transformational and transactional leadership had its origins in
the work of James McGregor Burns (Leithwood et al., 1992). Bass (1985a) extended
Burns’ initial work considerably during the mid 1980s. However, Bass et al.’s
(1985a; 1987) research work dealt mainly with non-educational institutions, and it is
only in recent years that transformational leadership theory has been seriously
applied to educational settings.
American Community Colleges—Roueche, Baker and Rose (1989) and
Murray and Feitler (1989). Roueche, Baker and Rose (1989), and Murray and
Feitler (1989) reported on studies designed to identify the impact of transformational
leaders on American Community Colleges. Roueche et al.’s (1989) study sought to
find validation for the constructs used in the model developed by Bass (1985a).
Roueche et al. (1989, p. vi) started by positing a definition of transformational
leadership, and argued it consisted of demonstrating “the ability to influence the
values, attitudes, beliefs, and behaviours of faculty and staff by working with and
through them to accomplish the college’s mission and purpose”. This research
focused on 1,220 presidents of junior and community colleges throughout the United
States. The study was undertaken in two phases: firstly, identification of
transformational community and junior college leaders; secondly, identifying those
practices used by those transformational leaders. Based on their analysis, Roueche et
al. (1989) identified five themes they believed were common to transformational
26
leaders. Transformational leaders: believe in teamwork and shared decision-making
(influence orientation); value people, both as members of a team and as individuals
(people orientation); understand motivation, and have high expectations of others
(motivation orientation); have a strong personal value system (values orientation);
and have a vision for what their institution can become (vision orientation). Included
among the transformational leadership practices identified in the study was the
concept of “shared” vision, which refers to the leader’s ability to clearly
communicate the college vision to subordinates, so that the leader’s vision becomes
the subordinate’s vision. Other transformational leadership behaviours that rated
highly included “encourages open communications” (influence orientation), “is
student centred” (people orientation), “assists in the development of others”
(motivation orientation) and “demonstrates a commitment to learning” (values
orientation).
Roueche et al.’s (1989) research verified the findings of the non-educational
sector regarding the central place of “vision” for the transformational leader (Bass,
1985a; Bennis & Nanus, 1985). Vision is “the distinguishing characteristic of the
transformational leader” (Roueche et al., 1989, p. 110). Vision allows the leader to
do more than simply know where their institution is heading. For the
transformational leader it adds depth and breadth to their understanding for the future
of their particular institution. It allows the leader to dream dreams, to imagine
outcomes, and to take risks proving the worthiness of their dreams. Finally, as
leaders communicate their vision with others, it becomes a shared dream with the
power to alter the institution and its environment profoundly (Roueche et al., 1989).
Murray and Feitler (1989) reported on the interaction of transactional and
transformational leadership styles on the organisational effectiveness of small,
private mid-western colleges in the United States. Data was obtained using two
questionnaires: Cameron’s (1978) The Assessment of Organisational Structures and
Effectiveness in Colleges and Universities (AOSECU) and Bass’ (1985a) Multifactor
Leadership Questionnaire—Form 5 (MLQ). The AOSECU was distributed to 15 key
administrators (academic, financial, student affairs, deans, heads of departments) in
each of the eleven institutions under investigation, while the MLQ was distributed to
“a person who reported to each of the persons completing the effectiveness
27
questionnaire” (Murray & Feitler, 1989, p. 9). A point of interest in this study was
the perceived differences reported between higher and lower level administrators.
Higher-level administrators perceived higher levels of inspirational leadership than
did lower level administrators. Murray and Feitler (1989, p. 20) suggested “the
possible desire for self-aggrandisement for those near the top or depreciation by
those lower down may explain a portion of the differences noted in this study”.
School Reform and Transformational Leadership—Leithwood and Jantzi
(1990). Leithwood and Jantzi (1990) completed a number of studies that examined
different aspects of transformational leadership style applied to schools. Leithwood
and Jantzi (1990) examined transformational leadership in relation to the reform of
school culture. Leithwood et al.’s (1990) research acknowledged that collaborative
school cultures have been associated with the implementation of a number of school
reform initiatives, including the empowerment of teachers. However, not much is
known “about (how) such cultures develop and whether or how school administrators
can facilitate that process” (Leithwood et al., 1990, p. 3). The transformational
leadership style was chosen to guide this study, as “transformational leadership,
while still vaguely specified, nevertheless was a promising conception of the type of
leadership required to meet many school reform objectives, not least the
development of a collaborative or shared, technical culture” (Leithwood et al., 1990,
p. 2). The aim of the study was to clarify conceptually the role of the
transformational leader, and to identify some of the specific strategies associated
with it that would facilitate the growth of a shared, technical culture. Leithwood and
Jantzi (1990, p. 9) commenced their study by considering transformational leadership
as a dependent variable, acknowledging it as a “primitive conception of potentially
useful leadership strategies” and collaborative culture as an independent variable.
Twelve schools were selected for inclusion in this study, with data collected in the
form of semi-structured questionnaires, the interviewer recording the interview in
each case. A form of content analysis was used to process the data collected, and a
causal network with an accompanying narrative was developed for each school.
Several clearly defined “principal strategies” or behaviours were identified
that were associated with the development of “shared technical staff cultures”
(Leithwood & Jantzi, 1990). Six strategies were highlighted, including:
28
1. strengthening school culture—by clarifying and prioritising the shared school
goals and reducing teacher isolation by creating opportunities for staff to
interact;
2. using bureaucratic mechanisms—by directing money for the allocation of
resources and setting up formal decision making structures;
3. staff development—by implementing a staff development program that
“acknowledges what can be learned from one’s immediate colleagues, as well
as others” (Leithwood & Jantzi, 1990, p. 26), and by principals directly
giving workshops in areas of their expertise;
4. direct and frequent communications—by making the school principal the
source of communications;
5. sharing power and responsibility with others—by establishing school
improvement teams which share responsibility; and
6. using symbols and rituals to express cultural values—by principals publicly
recognising the work of all staff and students which contributed to school
improvement.
Leithwood and Jantzi (1990) argued that their data provided support for the
notion that principals who are “transformational” have access to strategies that assist
in the development of collaborative school cultures. Further, Leithwood et al. (1990)
claimed that the transformational effect of the strategies identified may well be
explained in terms of altering the patterns of interaction among staff. This, they
suggested, was because “meaning” is socially constructed, and the meaning a
particular staff member brings to their work is a product of the schemata that they
possess in relation to their work. Changes in school culture to a shared, technical one
will involve changes in schemata held by individual staff members. Collaborative
cultures present “teachers with a different order of dissonance about purposes and
practices to which they must adapt their classroom schemata” (Leithwood & Jantzi,
1990, p. 31). Peers, including the principal, are more likely to stimulate a teacher to
consider more ambitious purposes and modifications to their teaching practice. Each
of the strategies identified in the study allowed principals the opportunity to clarify
explicitly “the preferred content of relevant schemata from their point of view”, or
allow for interaction between principal and staff on the content of schemata
29
(Leithwood & Jantzi, 1990, p. 32). Finally, sharing power may provide a stimulus for
developing shared meaning between principal and staff.
Principals’ Problem Solving Processes—Leithwood and Steinbach (1991).
Leithwood and Steinbach’s (1991) research focused on principals’ problem solving
processes. Specifically, their study was designed to answer two related questions:
What is the extent and nature of differences between expert and typical principals’
processes for solving problems with their staff? and: How do differences in
principals’ problem solving processes explain differences in their impact on schools?
Leithwood and Steinbach (1991) anticipated that a transformational leadership
construct would be helpful in identifying difference between expert and typical
principal style. They believed that expert principals, as opposed to typical principals,
used a group problem solving approach in school management. This, Leithwood and
Steinbach (1991, p. 222) asserted, was supported by previous research, where “group
as distinct from individual problem solving was increasingly valued by all principals
as they gained experience”. Leithwood and Steinbach (1991) suggested three reasons
why transformational leaders are more efficient at problem solving. Firstly, better
solutions can be achieved through collegial rationality. Secondly, the long-term
growth of participants can be achieved through the creation of a “zone of proximal
development”. These zones occur when the problem solving ability of the group
seems superior to the individual’s ability to solve problems, and individual
participants recognise this superiority. Lastly, transformational experience within
schools has the effect of increasing commitment to shared goals. The study used a
form of content analysis in which principals’ responses were coded at three points in
time: prior to a staff meeting, at the staff meeting itself, and at a post staff meeting
interview. Prior to this, principals were divided into two categories, “expert” and
“typical” principals, based on peer evaluation and extensive interview. Stimulated
recall methods in the context of staff meetings were used to generate the data on
problem solving processes. Expert and typical principals differed significantly in
their approach to “problem interpretation”. Typical principals revealed no sign of
conscious reflection on their own problem interpretation processes. Also, expert
principals varied considerably compared to typical principals with regard to the
account taken of others’ interpretations of the problems they were addressing.
Further, expert principals viewed each problem in the context of the larger school
30
mission. A particularly striking difference between expert and typical principals was
the degree of clarity they had about their interpretation of the problem, and their
ability to describe and justify their interpretation for their staff (Leithwood &
Steinbach, 1991). Typical principals were often unclear about their interpretations,
and experienced some difficulty in explaining their interpretations to others.
Leithwood and Steinbach (1991) noted that as far as “goals” were concerned, both
types of principals made a point to share their personal goals with those others
involved in the problem solving. However, a clear difference lay in the fact that
expert principals used the problem solving process to develop shared goals with their
staff. Typical principals were only concerned with achieving their own goals and
persuading others to agree with them about what those goals should be. Statistically,
where “values” were concerned, expert principals expressed about 17% more value
statements than did typical principals (Leithwood & Steinbach, 1991). Leithwood
and Steinbach (1991) also noted that expert principals were more able to anticipate
and cope with “constraints” or obstacles to problem solving than were typical
principals. Typical principals either failed to anticipate obstacles or identified
superficial obstacles. Expert principals, on the other hand, carefully planned in
advance how they would deal with anticipated constraints and responded flexibly and
adapted to unanticipated obstacles when they arose.
A major difference between expert and typical principals was found in the
area of solution processes. Typical principals rarely planned for collaborative
problem solving experiences with their staff (Leithwood & Steinbach, 1991).
Alternatively, expert principals would carefully develop plans that emphasise a
collaborative problem solving approach by providing clear introductions to problems
and their background. Typical principals invariably assumed that other staff members
had the same interpretation of the problem. Expert principals not only sought other
points of view, they were able to do so without intimidating or restraining their staff.
Leithwood and Steinbach (1991) argued that the differences in the problem solving
practices of expert and typical principals provide evidence of transformational
leadership behaviour. The author’s data suggested that “the everyday act of solving
problems in groups offers principals significant opportunities for exercising
transformational leadership but that typical principals do not make use of this
opportunity” (Leithwood & Steinbach, 1991, p. 239). Furthermore, the authors
31
suggested that expert principals could become even more transformational by
encouraging individual staff to be more reflective of their own problem solving
processes. Transformational principals will encourage collegial rationality by seeking
from their staff a broader range of perspectives from which to interpret problems and
in doing so, will be offered an expanded array of solutions. Also, transformational
principals will create a zone of proximal development, which reflects the attitude that
group solutions are superior to individual solutions. Finally, transformational
principals will create increased commitment to shared goals by increasing the
explicitness of the goals, developing moderately difficult goals and fostering
interaction about those goals. Central to these three constructs helpful in identifying
transformational leadership behaviour is what Leithwood and Steinbach (1991, p.
241) call “authentic dialogue”, where principals are open-minded, honest, careful,
attentive to group needs and attentive to their own thinking.
School Restructuring—Leithwood, Jantzi, Silins and Dart (1992).
Leithwood, Jantzi, Silins and Dart (1992) reported a study that examined the role
principals played in efforts of school restructuring. This study had two objectives; to
consider the processes that lead to restructuring, and to consider the consequences of
restructuring for those in a formal leadership role. Leithwood et al. (1992) identified
five variables that interacted together to form a model of school restructuring. These
variables included out-of-school processes (external environment influences that
affect educational policy development and implementation), in-school processes
(including school goals, teachers, leadership, culture, programs and instruction and
school policy, organisation and resources), organisational outcomes (staff perception
of the nature of change) and student outcomes (staff perception of the influence of
the change policy on student outcomes). School leadership was examined in terms of
transformational and transactional constructs. Out-of-school processes were
hypothesised to have direct effects on all the other constructs, while in-school
processes were hypothesised to have a direct effect on both the organisational and
student outcomes variables. It was supposed the variable of school leadership effects
had an indirect effect on student outcomes because of the mediation of in-school
processes and organisational outcomes. Both quantitative and qualitative methods
were used to address the questions posed in this study. Initially, a 162-item
instrument was developed to measure the model’s constructs. Data was collected
32
from 770 individuals across 272 schools. Six schools were designated as “lead”
schools and selected for case study. These “lead schools” had developed a reputation
for making progress with the implementation of a change program within the British
Colombia educational system (Leithwood et al., 1992). Teachers and principals were
interviewed at each of the selected schools, and a causal network was developed for
each school. Common patterns across cases were identified, and a “smoothed”
network was derived that reflected the relationship that appeared in each of the six
cases linking school leadership, in-school processes and organisational outcomes.
Three significant findings were reported. Firstly, the data gathered supported the
view that school leadership does not have a significant direct effect on student
outcomes and probably not on organisational outcomes (Leithwood et al., 1992).
Rather, the effect of school leadership on student outcomes is an indirect effect
mediated through the effect on teachers. This is consistent with recent theoretical
findings that explain leadership effects in schools and in other non-educational
organisations. Principal leadership behaviours have a greater direct effect on teachers
rather than on students. Secondly, school leadership had a significant direct effect on
in-school processes. Leithwood et al. (1992) identified “in-school” processes,
including school goals, school culture and policies and organisations as most directly
affected by the transformational role of the principal. School leadership influenced
teachers through the mediating variable of school culture, which in turn mediated
school goals. Finally, the effect of transformational school leadership behaviours on
in-school processes was demonstrated to be statistically more significant than the
effects of transactional leadership. Transactional leadership had little effect on school
restructuring programs, although transactional leadership behaviours implied a set of
managerial practices that are likely to be essential to the routine operation of a school
(Leithwood et al., 1992).
The results of Leithwood et al.’s (1990; 1991, 1992) earlier studies pointed to
the efficacy of transformational school leadership as opposed to transactional school
leadership, particularly when applied to the areas of steering school reform, problem
solving and restructuring. These results also indicated consistency with other school
effectiveness studies (Hallinger & Heck, 1998; Lezotte, 1991), which highlighted the
mediating indirect effect principal leadership plays in terms of meeting
organisational outcomes.
33
Principal Leadership in Private Schools—Hoover, Petrosko and Schulz
(1991). Hoover, Petrosko and Schulz’s (1991) study was designed to replicate Bass’s
(1985a) investigation, which examined military personnel and business leaders to
describe differences between transformational and transactional leadership styles.
However, Hoover et al. (1991) were interested in determining if Bass’ (1985a)
findings could be applied to the leadership styles of forty-five headmasters in private
secondary schools in the southeastern United States. Hoover et al. (1991) defined
transformational leaders as those who can motivate followers to do more than they
originally expected to do by raising followers’ levels of awareness, by getting
followers to transcend their own self interests, or by altering their need levels. A
transactional leader “recognises what the follower needs and clarifies for the follower
how these needs will be fulfilled in exchange for the follower’s satisfactory effort
and performance” (Hoover et al., 1991, p. 6). Hoover et al. (1991) used Bass’s
Multifactor Leadership Questionnaire Form 5, an instrument of 70 items consisting
of descriptive statements about superiors, which was distributed to the staff in the
forty-five schools that had agreed to participate in the study. A factor analysis of the
data yielded six significant factors. An examination of the rotated factor loadings
revealed six factors similar to those found by Bass and Hater (1988) in earlier
research, and these factors accounted for a total of 54% of the variance in the data.
The charisma factor loaded very highly and was unquestionably the most important
factor in explaining the variance in the data amongst the high school principals
surveyed. Many of the “charisma” items included statements regarding the mission
of the principal that is transmitted to staff, trust in the principal, the symbolic nature
of the principal’s leadership, the level respect engendered by the principal and
motivational abilities of the principal. Other factors identified in the data were found
to be consistent with Bass and Hater’s (1988) earlier study. These included
individualised consideration, intellectual stimulation and contingent reward. The
factor management by exception was divided into passive and active sub-
components. The passive factor “described leaders who wait for a failure before any
action is taken” (Hoover et al., 1991, p. 14), and included statements such as “the
leader is a firm believer in “if it ain’t broken, don’t fix it”“. Active management by
exception was characterised by items that described leaders who actively sought
irregularities so that corrective action could be taken. Hoover et al. (1991)
commented on the extra effort and inspiration factors of Bass and Hater’s (1988)
34
instrument. Extra effort, or the extent to which the leader elicited unusual effort from
subordinates did not load on any one factor, rather was scattered across all factors.
However, the items measuring inspiration tended to load on the charisma factor.
Hoover et al. (1991) interpreted this effect as being due to the notion that inspiration
is a subset of the charisma factor, and that inspirational leadership and charisma
cannot be separated.
Hoover et al. (1991) compared their results to the findings of two previous
comparative studies. While the results of all three studies were not completely
identical, all three studies were in substantial agreement (Hoover et al., 1991).
Hoover et al. (1991) concluded that their data supported the transformational and
transactional leadership constructs posited by Bass (1985a), even though their sample
size was considerably smaller than the two other reported studies. Further, they
suggested that the factor identified as “charisma” might be better designated
something else, as “so many items measuring constructs other than charisma were
part of this factor” (Hoover et al., 1991, p. 29). Hoover et al.’s (1991) study also
differed from other reported studies in that the factors charisma, intellectual
stimulation and individualised consideration clustered together in a higher order
analysis. Similarly, contingent reward and management by exception clustered
together. This, the authors argued, is evidence that the two separate factors of
transformational and transactional leadership behaviour could be identified as second
order factors. Lastly, Hoover et al.’s (1991) research confirmed Bass’ (1985a)
original findings that “satisfaction” and “effectiveness” were more highly correlated
with transformational leadership behaviour than they were with transactional
leadership behaviour.
Transformational and Transactional Leadership style behaviours in
Educational Organisations—Kirby, King and Paradise (1992). Kirby, King and
Paradise (1992) noted that the lack of research in the area of transformational
leadership in educational settings was the primary motivation for their study.
Conceptually, transformational leadership has great appeal. However, the authors
acknowledged that more refinement was needed before the concept can offer
something in the way of practical guidance to educational practitioners (Kirby et al.,
1992). Their research was an attempt to do this by investigating teacher’s perceptions
35
of the degree to which transformational leadership is practised by their immediate
supervisors. Kirby et al. (1992) used Bass’ (1985a) operationalisation of the
transactional and transformational leadership model in the course of their study. Bass
(1985a) concluded that both transactional and transformational styles are used by
business and military leaders, but that the relative use of these forms of leadership
will vary from leader to leader. Furthermore, Bass (1985a) “found that
transformational leadership significantly augments the power of transactional
leadership alone in predicting follower satisfaction and perceptions of leader
effectiveness” (Kirby et al., 1992, p. 304).
Kirby, King and Paradise (1992) reported on two studies designed to
determine firstly the degree to which educational leaders were perceived to use
transformational and transactional leadership behaviours, and secondly which
behaviours were best able to predict follower satisfaction and leader effectiveness. A
quantitative approach was taken in the design of the first study in which 103 teachers
were asked about their immediate superiors. Statistical analyses were applied,
consistent with procedures used by Bass et al. (1987; 1988) in earlier studies. The
constructs of charisma, individualised consideration, intellectual stimulation,
inspiration and contingent reward were significantly related to perceived
effectiveness and satisfaction with the leader. A stepwise regression procedure was
used to determine whether transformational leadership significantly augmented the
power of transactional leadership alone in predicting effectiveness and satisfaction.
Kirby et al. (1992, p. 305) noted that transformational leadership “was apparently
associated with higher levels of performance and satisfaction”. However, they also
noted that a closer investigation suggested two substantial problems with
interpretation of their data. Firstly, a forward regression analysis without forced entry
of the transactional subscales suggested that the charisma and laissez-faire constructs
were significant in predicting satisfaction, whereas charisma and intellectual
stimulation were significant predictors of perceived effectiveness. This
multicollinearity of subscales, Kirby et al., (1992, p. 305) noted, “creates difficulty in
interpreting the unique effects of each”. Secondly, the relation between the
dependent variables of perceived effectiveness and satisfaction confused
interpretation. Satisfaction and perceived effectiveness are two different outcomes;
yet, as was expected, the data revealed a strong relation.
36
Kirby, King and Paradise (1992) highlighted two further relevant problems.
Firstly, much research using the Multifactor Leadership Questionnaire (MLQ)
suffers from single subject bias, where one subject rates the leader on both
independent and dependent measures. To overcome this, Kirby et al. (1992)
suggested that confirmatory evidence from other studies using alternative dependent
measures is available (Kirby et al., 1992). Secondly, Kirby et al. (1992) noted that
the construct of charisma appears to overwhelm the other MLQ factors in explaining
perceived outcomes. This, they suggested, may be due to Bass’ (1985a) operational
definition of charisma rather than to any conceptual problems.
Kirby, King and Paradise (1992) also reported a second, qualitative study in
which a separate sample of nine teachers were asked to think of an extraordinary
leader in education with whom they had worked, and describe an event in which they
had participated that best exemplified that person’s leadership. Details surrounding
the event included how the event was initiated, who was involved, the objectives, the
leader’s actions and outcomes. After the narrative was finished, participants
completed an instrument which assessed the difficulty of identifying extraordinary
leaders in education, how effective the selected leader was in accomplishing their
goals, how satisfied employees were to work for this leader and to what extent their
selected leader was perceived to be unique. The nine narratives were analysed to
determine the behaviours and characteristics of extraordinary educational leaders. A
constant comparative analysis methodology was used to code the data. Bits of data
were sorted according to five themes: setting/event, goals, leader behaviour, leader
characteristics and outcomes. No specific categories within themes were established
a priori. Rather, categories within themes were created as the data was being coded.
A number of findings were supported by the data. Firstly, a leader’s ability to
inspire extra effort was associated with the specific behaviour of “modelling” (Kirby,
King & Paradise, 1992, p. 307). Modelling was viewed by most subjects as a
powerful form of persuasion, where leaders modelled the types of attitudes and
behaviours they expected to see in their staff. A second finding closely related to the
concept of modelling deals with the importance of communicating expectations and
challenging (Kirby, King & Paradise, 1992). Challenging followers was seen as an
accepted leadership strategy because leaders were viewed as risk takers. Leaders
37
challenge the status quo, but never recklessly. Challenges took the form of carefully
calculated change initiatives, and leaders took measures to ensure that these changes
had a “high probability of success” associated with them (Kirby, King & Paradise,
1992, p. 308). This included that leaders guaranteed success by enlisting the support
of “power brokers” to expand their sphere of influence. Leaders chose language
carefully and powerfully to gain support for their change initiatives. Slogans,
acronyms and symbols were used to build enthusiasm. Leaders were able to paint a
vision of a “big picture”, yet at the same time were able to give attention to minute
detail. Thirdly, extraordinary educational leaders supported change by providing
opportunities for training and development. These leaders saw staff development as
opportunities for growth. Often, leaders would provide staff development
opportunities for a small group of staff, who would in turn be given the responsibility
for others, until entire school communities were initiated. Lastly, “rewards seemed to
be de-emphasised by extraordinary school leaders” (Kirby, King & Paradise, 1992, p.
309). Kirby et al. (1992) suggested that while this de-emphasis may have resulted
from a lack of material resources from which to reward, extraordinary educational
leaders seemed to place a strong belief in the power of intrinsic reward.
Kirby et al. (1992) concluded that while there were extraordinary or
transformational leaders in education, teachers preferred a leader who mixed
transformational leadership behaviours with the high order transactional behaviour of
contingent reward. Although these extraordinary leaders are difficult to find,
participants who could identify them had no trouble in describing examples of that
person’s leadership or the specific attitudes and behaviours that made that leader
extraordinary (Kirby, King & Paradise, 1992). They surmised that “followers prefer
leaders who engage in transformational behaviour associated with individualised
consideration, intellectual stimulation and the transactional behaviour of contingent
reward” (Kirby, King & Paradise, 1992, p. 309).
Kirby et al. (1992) supported the findings of other researchers (Sashkin,
1988), that transformational leaders use effective communications to paint visions of
preferred futures. Furthermore, transformational leaders are consistent in their
actions to support their vision, value and respect the input of others, and act to create
risks that others can support and achieve.
38
While Kirby et al. (1992) confirmed Bass’ (1985a) findings that certain
leadership behaviours are necessary to elicit satisfactory performance and other
leadership behaviours enhance performance to levels beyond expectations, they do
not agree that ordinary leadership is “entirely transactional”. Many of Kirby et al.’s
(1992) respondents saw structuring activities (for example, providing resources and
selecting key participants) as a necessary prelude to extraordinary accomplishments.
They concluded that extraordinary leaders are careful not to overdefine structure, and
that they monitor structural development so that performance is optimised.
Transformational Leadership in Inclusive Schools—Ingram (1997). Ingram
(1997) studied the effectiveness of principals as educational leaders in inclusive
schools in the United States. Inclusive schools are those where both regular students
and disadvantaged students learn together in the same classrooms. Their success or
otherwise, argued Ingram (1997), is largely determined by the leadership behaviours
demonstrated by the principal. The fundamental question raised in this research,
therefore, was this: are principals more likely to demonstrate transformational
leadership behaviours to a greater degree than they demonstrate transactional
leadership behaviours? Ingram (1997) gathered data from forty-four teachers across
15 schools who worked in the Michigan Inclusive Education Project over five school
districts. The target teachers were asked to complete Bass and Avolio’s (1990)
Multifactor Leadership Questionnaire (MLQ Form 5R), a 70-item form designed to
measure seven leadership behaviours (four transformational, two transactional and
one non-leadership behaviour) along the transformational and transactional
continuum. The transformational leadership behaviours included charisma,
inspiration, intellectual stimulation and individualised consideration, while the
transactional leadership behaviours were defined as contingent reward and
management by exception. The non-leadership factor was laissez-faire leadership.
The MLQ also included items that measured follower satisfaction with leadership,
extra effort and overall leader effectiveness. Reliability analysis on each of the
leadership factor scales produced acceptable scores of between .75 and .94, except on
the scales of contingent reward (.68) and management by exception (.59). In the
original study, Ingram (1997) used Bass and Avolio’s (1990) operational guidelines
to categorise the leadership behaviours into three groups: high transformational/low
transactional, low transformational/high transactional and moderate
39
transformational/moderate transactional leadership styles. However, multiple
regression analysis of the categories revealed only two groupings: high
transformational/moderate to low transactional, and high transactional/moderate to
low transformational leadership style behaviours. Ingram (1997) found statistically
significant differences between the means of the two groups under investigation,
supporting the hypothesis that principals in inclusive schools are perceived by
teachers to demonstrate greater degrees of transformational behaviour than
transactional behaviour. Further, an analysis of the three outcome measures
(satisfaction, extra effort and effectiveness) demonstrated that principals who were
perceived to display a high degree of transformational leadership behaviour have a
greater impact on teacher’s motivation to exert extra effort than principals who
display transactional behaviours. Principals who displayed transactional leadership
behaviours only moderately influenced teachers’ motivation. This, Ingram (1997)
argued, supported Bass’ (1985) original contention that transformational leadership
behaviours augment transactional leadership behaviours. Overall, Ingram (1997)
found evidence to support the contention that transformational leadership behaviours
are more effective than transactional leadership behaviours in inclusive schools. The
transformational leader more than likely will emphasise the development of shared
values and beliefs, and commitment to common goals, whereas the transactional
leader will more than likely emphasise control aspects of administration,
enforcement of district policies and stabilisation of programs and teacher
competency.
Principals’ Leadership Qualities—Leithwood and Jantzi (1997). Leithwood
and Jantzi (1997) examined the influences that caused teachers to attribute leadership
qualities to their principals. This, they argued (Leithwood & Jantzi, 1997) is an
important question for principals if they are to exercise the type of influence that
most of them aspire to. Further, the transformational leadership paradigm provided a
good platform from which to examine leadership practices in schools, because of its
high potential to build commitment among followers in times of turbulence in
educational environments. In particular, the transformational leadership behaviours
examined by Leithwood and Jantzi (1997) included fostering vision and goals,
development of a collaborative decision-making structure, the promotion of good,
professional practice, providing individualised support, providing intellectual
40
stimulation and holding high performance expectations. Leithwood and Jantzi (1997)
sought to compare these behaviours against those behaviours they considered good
school managers need to exercise. These management behaviours included
establishing effective staffing practices, the provision of instructional support,
monitoring school activities and provision of a community focus. Data for this study
was obtained from a 243-item instrument specifically developed by Leithwood and
Jantzi (1997) to measure transformational leadership and school management
constructs collected from the teachers and principals in one large school district in
Ontario, Canada. The intended sample size was originally designed to be 2,378
teachers (1,632 teachers from 100 elementary schools, and 746 teachers in 15
secondary schools), but missing data reduced this sample to 1,253 teachers in total.
Reliability analysis resulted in Cronbach alphas ranging from between .93 and .87
for the transformational leadership constructs, and .92 to .76 for the four school
management factors. The results reported by Leithwood and Jantzi (1997) provided
an interesting insight into how teachers’ perceptions of leadership behaviour evolve.
Leadership, they argued (Leithwood et al., 1997) is an influence process that is
dependent on a followers consent to be led. Therefore, being perceived as a leader is
just as critical as being effective as a leader.
This study has important theoretical and practical implications for “would be”
leaders. Doing good work and being seen as doing good work is a powerful strategy
which positively influences teachers’ perceptions of leadership (Leithwood & Jantzi,
1997). As far as teacher perception of transformational leadership is concerned, what
the principal does (actions and perceived effects of actions) is more important than
who the principal is (age and gender). The most powerful variable explaining
teachers’ leader perceptions, argued Leithwood and Jantzi (1997) is the leader’s
contribution to those in-school dimensions of the school’s mission and goals, culture,
structure and organisation, policies and procedures, planning and instruction. Visibly
contributing to these in-school dimensions in ways that teachers find helpful is likely
to be interpreted as a sign of transformational leadership.
Leadership and Student Engagement—Leithwood and Jantzi (2000).
Leithwood and Jantzi (2000) have argued that transformational leadership practices
are conducive to the development within teachers of increased capacity and
41
commitment for change. This extra capacity and commitment is assumed, in turn, to
result in extra effort and greater productivity (Leithwood & Jantzi, 2000). Leithwood
and Jantzi’s (2000) study was designed to examine the effects of transformational
leadership practices on organisational conditions and on student engagement with
school, taking into consideration the effects of family educational culture. Of
particular note in this study was the recognition that leadership effects will, at best,
be mediated through the influence of teacher’s classroom practices. In this study,
Leithwood et al. (2000) defined six transformational leadership behaviours or
practices as vision building, providing intellectual stimulation, offering
individualised support, symbolising professional practices and values, demonstrating
high performance expectations, and the development of structures that foster
involvement in decision-making.
Transactional practices, on the other hand, were those managerial practices
that were fundamental to organisational stability (Leithwood et al., 2000), and
included staffing practices, instructional support, monitoring school activities and
community focus. Four school organisational conditions were proposed, including
purposes and goals, school structure and social networks, people, and organisational
culture. Finally, two classroom conditions were also included in the study;
instructional services (defined as interventions by teachers designed to stimulate
educational growth) and policies and procedures (defined as guidelines for decision-
making within schools). Data was collected from 2,465 classroom teachers and
44,290 students in a school district in eastern Canada. Two instruments were used;
one designed to collect data from teachers on school conditions and transformational
leadership behaviours of principals, and the other instrument to collect data from
students on their engagement with school and family educational cultures. Leithwood
and Jantzi (2000, p. 124) concluded that transformational leadership practices had
strong and statistically significant direct effects on organisational conditions (such as
purposes and goals and school structure), but weak, although significant indirect
effects on student participation. While this is an obvious result, given that
organisational conditions are those facets of schools that transformational principals
have direct access to, it is these factors, according to Leithwood et al. (2000, p. 125)
that are “the means through which school effects are exercised”.
42
Transformation Leadership and Teacher’s Commitment to Extra Effort—
Geijsel, Sleegers, Leithwood and Jantzi (2002). Geijsel, Sleegers, Leithwood and
Jantzi (2002) reported the findings of a study designed to examine the effects of
transformational leadership styles on the commitment of teachers to extra effort in
two separate countries, Holland and Canada. This study of transformational
leadership style was based on the theoretical work of Burns (1978) and Bass (1985a),
but was extended by Leithwood to incorporate differences found within school
contexts. Specifically, transformational leadership style is defined by three practices,
namely: setting directions, developing people, and redesigning the organisation.
Within these three practices are ten constructs identified by Geijsel et al. (2002),
which resemble the constructs previously developed by Bass (1985a), including
vision building, individualised consideration and intellectual stimulation. Two
constructs were used to measure teacher commitment to change: capacity beliefs, or
the extent to which teachers feel uncertainty in their own capability in response to
daily pressures to reform and implement innovation; and context beliefs, or the
perceived extent of collegial and school structure support. Data was gathered in the
two countries using a battery of different but complementary instruments. In
Holland, 1,347 teachers from 45 schools responded to the survey, while in Canada,
1,444 teachers from 43 junior high and high schools completed the survey
instruments. The survey also gathered information about the demographic
characteristics of respondents, including respondent’s age, length of teaching service
and gender. Statistical analysis was undertaken using confirmatory factor analysis to
develop adjusted measurement models, then structural equation models were used to
describe both the direct and indirect relationships between the variables under
consideration. Of the findings reported in this study, Geijsel et al. (2002) found
modest evidence for the effect of transformational leadership on teacher’s
commitment to change in both countries. Specifically, transformational leadership
behaviours (vision building, individualised consideration and intellectual
stimulation) had more statistically significant effects on teacher’s context beliefs than
on capacity beliefs. However, within these overall results, it was found that vision
building and intellectual stimulation had more effect on context beliefs than did
individualised consideration.
43
Summary
Research in the area of transformational school leadership over the past
decade has pointed to the direct effects that a transformational principal can exercise
on the in-school aspects of their organisations, including mission and goal formation,
school culture, structure and organisation and the development of policies and
procedures (Leithwood & Jantzi, 1990). Results have also pointed to the differential
effect a transformational principal can have over a transactional leader in terms of
cultural reform (Leithwood et al., 1990), problem solving (Leithwood & Steinbach,
1991), school restructuring (Leithwood, Jantzi, Silins & Dart, 1992) and encouraging
extra effort from teachers (Kirby, King & Paradise, 1992). Further, researchers have
highlighted some transformational behaviours that principals can employ to influence
the in-school processes in their schools. These include sharing power in decision
making, recognising the potential of collaborative staff development (Leithwood &
Jantzi, 1990) and modelling those behaviours the leader desires to see developing in
their schools (Kirby et al., 1992). Finally, many studies have pointed to the
differential effectiveness of transformational school leadership over transactional
school leadership (Ingram, 1997; Leithwood et al., 1992; Leithwood et al., 1997).
Leadership Constructs Used in this Thesis
Bass (1985a) outlined a theoretical framework that differentiated between
transactional and transformational leadership and which defined a transactional
leader as one who recognises what followers want to get from their work, and tries to
see that this happens if performance warrants it. A transactional leader exchanges
rewards and promises of reward for performance. Bass (1985a) argued that a
transactional leader is responsive to the immediate self-interests of followers if they
can be met in the process of followers completing the work assigned to them. By
contrast, Bass (1985a) defined a transformational leader as one who motivates
followers to do more than they originally expected to do. Transformational leaders
can achieve this in three ways. Firstly, a transformational leader raises the level of
follower awareness and consciousness regarding the importance and value of
designated outcomes, and ensures ways of reaching those outcomes. Secondly,
44
transformational leaders encourage followers to transcend their own self-interest for
the sake of the team or organisation. Thirdly, transformational leaders alter follower
needs by expanding their portfolio of needs and wants.
Bass’ (1985a; Bass & Avolio, 1997) theory is centred on the high-order
changes in both the effort and performance of followers that accompanies a
transformational style of leadership, while traditional leadership theory focuses on
first-order changes. Bass (1985a) argued that there is a “certain kind of leader who is
capable of going beyond first-order changes to high-order change and who inspires
people to heights they never intended to achieve” (Hoover, Petrosko & Schulz, 1991,
p. 2). Bass (1985a) labelled this type of leader as transformational and argued that
follower motivation to work cannot be completely accounted for using a
transactional approach to leadership. While the “transaction” is important (that is, the
exchange of needs for needs) and is even apparent in all forms of leadership
interaction, it fails to account for the “considerable portion of the relationship
between leaders and subordinates” (Hoover et al., 1991, p. 3). He contended that
psychological theory provided a basis for explaining a broader conceptualisation of
leadership. Using Maslow’s theory of human motivation, Bass (1985a) argued that
some leaders are able to motivate followers far beyond the lower level on Maslow’s
hierarchy into the self-actualisation level. Managers, he claimed, operate at low
levels on Maslow’s hierarchy and are impersonal, relating only to people according
to their organisational roles. Leaders on the other hand, are personal and active,
relating to others empathetically and intuitively while projecting their ideas and
visions on to their organisational landscapes. Bass’ (1985a) transformational leader
“is not satisfied with meeting some minimum standard, which could become the
maximum” (Hoover et al., 1991, p. 4).
Transformational leaders encourage followers to develop and work beyond
the normal level of expectation (Bass & Avolio, 1997). Followers do this for several
reasons, including personal identification with the efforts of the leader in making
sacrifices to achieve the stated outcome. Bass et al. (1997) argued that this process
raises follower motivation, enhances self-efficacy and encourages willingness on the
follower’s part to accept extraordinary challenges. Fundamental to Bass’ (1985a;
Bass & Avolio, 1997) conceptualisation of leadership is that transformational
45
leadership behaviour augments or enhances the effects of transactional leadership
behaviour in predicting satisfaction and other outcome effects on followers. This,
Bass and Avolio argue, is an important principle of the original model, namely, that:
Transactional leadership provides the basis for effective leadership, but a greater amount of extra effort, effectiveness and satisfaction is possible from employees by augmenting transactional leadership with transformational leadership. The full range of potential is achieved through both, not either one versus the other style of leadership. (Bass & Avolio, 1997, p. 22).
Bass’ (1985a) original operationalisation of the transactional and
transformational model of leadership has undergone a process of refinement and
consolidation since the time it was first developed. The model, as used in this thesis,
includes five transformational constructs, four transactional constructs, one non-
leadership construct and three outcome constructs (Bass et al., 1997).
The transformational constructs (Bass & Avolio, 1997) used in this thesis are:
1. Idealised Influence (Charisma)—Attributed: this refers to a follower’s desire
and willingness to embrace the vision and mission as explained by the leader.
2. Idealised Influence (Charisma)—Behaviour: this refers to a follower’s desire
to emulate and identify with the leader. Leaders are trusted and highly
respected, holding much referent power, and set challenging yet attainable
missions for their followers;
3. Inspirational Motivation: the degree to which the leader raises followers’
enthusiasm and respect, and inspires loyalty, by using symbols, metaphors
and simplified emotional appeals to increase awareness of a mutually desired
goal;
4. Intellectual Stimulation: the degree to which the leader provides intellectual
and problem-orientated guidance. Followers are encouraged to think on their
own, and consider creative ways to attain group goals; and
5. Individualised Consideration: the degree to which the leader is concerned
with the individual needs of followers (for example, mentoring). The leader
delegates in order to stimulate and create learning experiences, and treats
each person with respect. (Bass & Avolio, 1997).
46
The transactional constructs (Bass & Avolio, 1997) referred to in this thesis
are:
6. Contingent Reward: the degree to which the leader makes clear what the
follower must accomplish in order to be rewarded;
7. Management by Exception—Active: the degree to which the leader monitors
follower actions to ensure no mistakes are made, and so allows the status quo
to continue;
8. Management by Exception—Passive: the leader intervenes only to make
corrections when things go wrong.
There is a possibility that neither transformational or transaction leadership
will be evident in the leader/follower relationship. Bass et al. (1997) referred to this
as non-leadership. The non-leadership construct (Bass & Avolio, 1997) used in this
study is:
9. Laissez-Faire: this indicates the absence of leadership and the avoidance of
intervention. In this case, the leader avoids making decisions, delays in giving
feedback or reward, and is often absent and not involved in meeting group
objectives.
Leadership Summary
The application of leadership theory and research has yielded mixed results as
far as understanding leadership processes in schools is concerned. New
transformational paradigms have the capability to extend our understanding of the
processes involved. By allowing researchers to consider the separate factors that are
involved in this process, the promise is that differentiated or targeted outcomes can
be achieved. The current investigation capitalised on Bass et al.’s (1997)
transformational leadership theory and research in order to operationally define the
multiple domains of transformational leadership, transactional leadership and the
non-leadership construct of laissez-faire leadership, to examine the influences of the
principal’s leadership behaviour on school learning environment and selected teacher
outcomes.
47
School Learning Environment
Schools have come under increasing pressure to improve their educational
performance, with much of the “effective school” literature emphasising the need to
“strive for excellence” (Owens, 1998, p. 92; Walker, 1990). Clearly, effective
schools have effective leaders, who are focused on attaining the goals of their
organisations. Effective leaders in turn are perceived positively by staff, students and
parents. This perception, held by the stakeholders of the school community
constitutes a school’s climate or learning environment (Keefe & Kelley, 1990). Lake
(1991, p. 1) argued that an effective school climate is one where “a positive attitude
on the part of the entire staff and student body (is) exhibited through their overt
behaviour that creates a warm, orderly learning environment”.
The Historical Development of School Learning Environment
Research
School Effectiveness and School Learning Environment
Research has consistently stressed the importance of school climate as a
variable influencing school effectiveness (Duignan, 1986; Eicholtz, 1984; Hoy &
Ferguson, 1989; Hoy & Miskel, 1996) and highlighted the principal’s role in setting
and manipulating school climate (Brady, 1988; Duignan, 1986; Hall, 1987; Hall &
George, 1999; Hoy & Miskel, 1996). However, after many years of substantial
empirical investigation, “the meaning of organisational climate remains elusive”
(Hoy, Tarter & Bliss, 1990, p. 260).
Generally, school climate or school learning environment can be regarded as
the feel or personality that each school exudes; some schools are perceived as good
schools, while others are not. Hoy and Miskel (1996) argued that school climate or
school learning environment is that enduring quality of a school’s environment
experienced by staff and affecting their behaviour. It is this subtle spirit of a school
that has generally been referred to as school morale. However, over the past twenty-
five years, it has been increasingly referred to as “school climate”.
48
Silver (1983, p. 180) defined school climate as “the tone, ambience or
atmosphere of an organisation—the sense that a place has a quality all of its own.
Halpin (Hoy & Miskel, 1996) regarded school climate as the organisational
personality of a school. Hoy and Miskel (1996) agreed, viewing school climate as
those sets of internal characteristics that distinguish one school from another and
influence the behaviours of their members. Keefe and Kelley (1990) defined school
climate as a relatively enduring pattern of shared perceptions about the
characteristics of an organisation and its members.
It would appear that no standard definition of school climate exists, and
conceptually it is regarded by many researchers as “complex and vague” (Hoy,
Tarter & Bliss, 1990, p. 260). Moreover, recent attention to school culture has only
added further to the confusion, as there too, no general agreement exists concerning
the difference between climate and culture. Furthermore, researchers (Anderson,
1982; Freiberg & Stein, 1999; Lindelow, Mazzarella, Scott, Ellis & Smith, 1989;
Rentoul & Fraser, 1983) reported that school climate can be measured at several
levels, including the classroom, the school and the district levels.
Given the wide range of perceptions, Anderson’s comment (1982) that the
field of climate research is in many ways reminiscent of the seven blind men who
gave seven different descriptions of an elephant based on what each man touched, is
understandable. All claimed to possess the definitive image of the elephant. Some,
said Anderson (1982, p. 371), see school climate as an albatross, “a burden to
policymakers who need information on mechanisms that can be easily manipulated
to affect student outcomes”, yet lack power because of the scant empirical evidence
on how to manipulate it. Others see climate likened to a unicorn, “a desirable focus
of study, but one which is unattainable” (Anderson 1982, p. 371). Finally, she
argued, some see climate research as likened to a phoenix, capable of being “born of
the ashes of past school effects research” (Anderson 1982, p. 372) which, however,
relied on poor modelling, inadequate measures, missing and wrong variables.
Purkey and Smith (1983) largely equated culture with climate. Kelley (1989)
defined culture as the characteristics and traditions of the school, and climate as the
relatively permanent perceptions of the characteristics of the school, that is, the
49
perceptions of the culture of the school. Hoy, Tarter and Bliss (1990) suggested that
culture consists of the shared assumptions, values or norms, whereas climate is
defined by shared perceptions of behaviour. Clearly, there is not a large conceptual
leap from shared assumptions or beliefs (culture) to shared perceptions of behaviour
(climate); however, the difference seems real enough (Hoy, Tarter & Bliss, 1990).
In this thesis, school culture and climate are seen as related. The difference
between them, according to Freiberg and Stein (1999) is in how they are viewed;
school learning environment is viewed from a psychological perspective, and school
culture is viewed from an anthropological perspective.
The question that has dominated school learning environment literature is that
of the relationship between climate and culture. Both have similar characteristics,
yet, to researchers, the differences are real enough. School climate or learning
environment is “more interpersonal in tone and substance and is manifest in the
attitudes and behaviours of teachers, supervisors, students, and principals at work”
(Sergiovanni, 1991, p. 218). School culture is a more normative concept in the sense
that it is a reflection of shared beliefs and values across a school community that
extends beyond interpersonal life. It is what the school stands for and believes about
education, and what it seeks to accomplish. Also, it includes the image the school
projects into the community.
Therefore, school culture refers to the beliefs, values and customs of a school
(Maxwell & Thomas, 1991) while school climate, or school learning environment
refers to the enduring perceptions of the characteristics of the school (Kelley, 1989).
School climate or school learning environment is that set of internal characteristics
that distinguishes one school from another and influences the behaviour of its
members, both staff and students.
Despite the problems associated with the definition of school climate or
learning environment in the literature, most writers agree that schools do possess
something called a learning environment that is unique to each organisation. Further,
each school-learning environment displays differences which, while discernible, are
elusive, complex and difficult to describe and measure. School learning
50
environments are influenced by certain dimensions of the school, and affect many
student and staff outcomes. Lastly, an understanding of the influence of school
climate or learning environment will improve the understanding and prediction of
student and staff behaviour (Anderson, 1982).
School learning environment is a complex concept that is influenced by a
wide range of variables. Anderson (1982) used a schema developed in the 1960s to
classify the variables identified by the research as influencing organisational climate
or learning environment. In it, climate is seen as a summary concept that is
concerned with the total environmental quality of an organisation. Tagiuri (in
Anderson, 1982; Miskel & Owaga 1988) described four environmental dimensions
that are identified as being associated with school climate:
• ecology—the physical and material aspects of a school;
• milieu—the social aspects consisting of particular persons and groups;
• social systems—the aspects concerning the patterns of group interactions; and
• culture—aspects of belief systems, values and cognitive structures.
Researchers Anderson (1982) and Miskel and Owaga (1988) employed this
taxonomy to review the literature on school climate from the early 1960s. Further,
they cited other researchers (James & Jones, 1974), who identified three
measurement approaches to reviewing organisational learning environment. The first
is a multiple measurement-organisational attribute approach, where school climate is
treated as a set of organisational attributes that are measured using a variety of
measurement scales. The second is a perceptual measurement-organisational attribute
approach, in which school climate is treated as a set of perceived variables that are
seen as organisational main effects. The third approach is a perceptual measurement-
individual attribute approach, where climate is treated as a perceived attribute of the
individual members of an organisation (Miskel & Owaga, 1988). While Tagiuri’s
framework provides an ontological basis for examining the climate literature, James
and Jones’ framework “can serve to identify epistemological patterns” (Miskel &
Owaga, 1988, p. 290).
51
Measuring School Learning Environment
School learning environment instruments can be useful tools for comparing
school climates over many schools, and for measuring changes in school climate
over time. However, all learning environment instruments have limitations, as they
cannot directly measure what is going on in a school. Rather, they measure
respondents’ perceptions of what is going on. Although this is a recognised limitation
of climate instruments (Lindelow et al., 1989, p. 174), it is not a “fatal flaw”. With
this limitation in mind, the following approaches have been developed to measure
school learning environment or climate. While this is not an exhaustive list of
instrument development, it does represent some major developments in the area of
school climate research.
The Behavioural Approach to Describing School Learning Environment
One of the earliest school climate measures was developed by Halpin and
Croft (1963), and was known as the Organisational Climate Description
Questionnaire (OCDQ). It focused on the social interaction that occurred between
teachers and principals. Halpin and Croft (1963) identified eight dimensions of
school climate related to teacher and principal behaviour. They included:
a. Teacher Behaviour:
1. disengagement—refers to the teachers’ tendency to go through the motions
without actual commitment to the task at hand;
2. esprit—refers to morale growing out of a sense of task accomplishment and
social needs satisfaction;
3. hindrance—refers to the teachers’ feeling that the principal burdens them
with routines and unnecessary paperwork; and
4. intimacy—refers to the teachers’ enjoyment of warmth and friendly personal
relations with one another.
b. Principal Behaviour:
5. aloofness—refers to formal and impersonal principal behaviour, where the
principal “goes by the book”;
52
6. production emphasis—refers to close supervisory behaviour on the part of the
principal;
7. thrust—refers to dynamic principal behaviour in which an attempt to move
the organisation is made through the example that the principal sets; and
8. consideration—refers to the warm, friendly behaviour of the principal, where
they try to be helpful and do extra things for the faculty when possible (Hoy
& Miskel, 1987).
Six school climates were then identified, based on the relationships between these
eight aspects of interaction. These climate types were arranged along a continuum,
and described as open, autonomous, controlled, familiar, paternal or closed. An open
climate was considered to be the most desirable, as it demonstrated a high degree of
thrust and esprit (sense of collegiality or community) and was low in disengagement,
while a closed climate was the least desirable of the climate types (Freiberg, 1998;
Hoy & Miskel, 1996). A closed climate, at the other extreme of the continuum, was
characterised by low thrust and esprit and high disengagement. Principal and
teachers simply go through the motions, the principal stressing routine and
unnecessary busywork. Teachers respond with minimal effort and little resultant
satisfaction. The principal’s ineffective leadership is seen in close supervision,
formality, impersonality and a lack of consideration for the faculty. The principal is
unable to provide dynamic, personal leadership, and misguided tactics produce
teacher frustration and apathy. Halpin and Croft (1963) defined their school climate
taxonomy as being comprehensive, as all schools would be included somewhere
along an open-closed continuum. Each of the six climate types along the continuum
was regarded as an ideal type and was situated in a specific position. However, in
reality, most school learning environments deviated in some way from the pure
climate types described by Halpin and Croft (1963). As a classification scheme,
nonetheless, this framework was useful in categorising schools and had been used in
several empirical studies (Silver, 1983).
Silver (1983) identified several problems associated with this
conceptualisation. It was a cumbersome framework that lacked a clear underlying
logic and the notions of openness and closedness were vague, lacking the sharp
precision that theory demands. Kottkamp, Mulhern and Hoy (1987) noted that use of
53
the Organisational Climate Description Questionnaire as a climate measuring
instrument is dated and limited to primary schools only. They also pointed out that
serious questions persist about the reliability and validity of at least half the subtest
scales (Kottkamp et al., 1987). However, Silver (1983) argued that although this
framework has weaknesses, this does not mean it should be done away with. In fact,
the behavioural description approach measuring school learning environment is one
of the few measurements that pertain specifically to schools rather than to
organisations in general. In this respect it is uniquely applicable to educational
systems. Its focus is on observable behaviours and, furthermore, it is linked to a
manageable instrument that is easily administered and quickly completed by
respondents (Silver, 1983).
Perceived weakness in the Organisational Climate Description Questionnaire
format has led to the development of other measures of climate since Halpin and
Croft (1963) first trialled the Organisational Climate Description Questionnaire (Hoy
& Miskel, 1996). Indeed, the Organisational Climate Description Questionnaire has
been revised several times, both for primary schools (Hoy & Clover, 1986) and for
secondary schools (Kottkamp, Mulhern & Hoy, 1987).
The Organisational Dynamics Approach to Describing School Learning
Environment
The organisational health of a school is a second way of conceptualising the
learning environment of a school. Miles (Hoy & Miskel, 1987, p.237) defined a
healthy organisation as one that “not only survives in its environment, but continues
to cope adequately over the long haul, and continuously develops and extends its
surviving and coping abilities”. Hoy and Feldman (1999, p. 87) agreed, defining a
healthy school as “one in which the technical, managerial, and institutional levels are
in harmony: and the school meets both its instrumental and expressive needs as it
successfully copes with disruptive external forces and directs its energies toward its
mission”.
Healthy organisations deal successfully with forces external to an
organisation, which tend to be disruptive, and direct their energies towards the goals
54
or mission of the organisation. The organisational health of a school has been defined
using seven specific patterns of interaction within a school. The Organisational
Health Inventory (OHI) measures the instrumental needs of adaptation and goal
achievement, and the expressive needs of social and normative integration. As well,
it does this at the three levels of responsibility and control within a school, namely, at
the institutional level, the managerial level and the technical level (Hoy & Miskel,
1996). Hoy and Miskel (1996) characterised a healthy school as one that is protected
from unreasonable community and parental pressures. The principal provides
dynamic leadership that is both task-oriented and relation-oriented. This behaviour is
supportive of teachers, and yet provides direction and maintains a high standard of
performance. Teachers are committed to teaching and learning, setting high but
achievable goals for students in an environment that is orderly and serious. In a
healthy school, teachers like and trust each other, are enthusiastic about teaching and
are proud of their school. Unhealthy schools are “vulnerable to destructive outside
forces”, and are bombarded with unreasonable demands from community and
parental groups (Hoy & Feldman, 1999, p. 91). The principal lacks clear leadership;
there is little direction and limited consideration and support for the teaching staff.
Teacher morale is low, and they act aloof, suspicious and defensive. The pressure for
academic excellence is low, as both principal and faculty are simply “putting in
time” (Hoy & Miskel, 1996, p. 155).
The Organisational Health Inventory is a relatively new instrument, and
consequently research support is limited. However, it is a useful tool as its
conceptual underpinnings are consistent with many of the characteristics of effective
schools. These include an orderly and serious environment, high yet attainable
student goals and a cohesive unit based on trust and dynamic principal leadership,
where principals blend their behaviour to fit the situation (Hoy & Miskel, 1996).
Pupil Control Approaches to School Learning Environment: Custodial to
Humanistic Controls
An alternative way of conceptualising the social climate of a school is in
terms of the dominant control patterns that principals and teachers use to control
students. Willower and Jones (in Anderson, 1982) developed an instrument called the
55
Pupil Control Ideology (PCI), based on field studies of the social systems of high
schools. Willower and Jones (in Anderson, 1982, p. 376) “established the constructs
of pupil control orientations as a school climate descriptor with direct implications
for pupil and teacher behaviour”. The Pupil Control Ideology measured teacher
orientation towards pupils on a humanistic-custodial continuum. A humanistic school
is “conceived of as an educational community in which students learn through
cooperative interaction and experience” (Hoy & Miskel, 1996, p. 158). Self-
discipline is high, while strict teacher control is low. The school atmosphere is
democratic, with open channels of two-way communication between staff and pupils.
A custodial school is regarded as the traditional school, where students are provided
with a rigid and highly controlled environment in which the maintenance of order is
primary. Teachers perceive the school as being autocratic, with the flow of power
and communications unilateral and downward. Teachers do not attempt to
understand student behaviour, they perceive students as both irresponsible and
undisciplined and believe they must be controlled through punitive sanctions (Hoy &
Miskel, 1996). A companion instrument, Pupil Control Behaviour (PCB), is also
often administered to students to describe their teachers. Hence, both the Pupil
Control Ideology (PCI) and Pupil Control Behaviour (PCB) instruments represented
a unique thread in school climate research, having both anthropological and
psychological roots (Anderson, 1982).
The National Association of Secondary School Principals’ School Climate Survey
The National Association of Secondary School Principal’s (NASSP) School
Climate Survey developed out of a desire to provide “a set of school climate
measures that would have psychometric validation and also be useful to
practitioners” (Lindelow et al., 1989, p. 172). A task force group led by James
Keefe developed several instruments in a battery called the Comprehensive
Assessment of School Environments (CASE), and included instruments designed to
measure school climate and teacher, parent and student satisfaction (Keefe & Kelley,
1990; Kelley, 1989). The CASE model was designed to encompass a wide variety of
inputs and outputs associated with the process of school improvement. At a broader
level, the model included the cultural setting of the school educational program. At a
district level, three areas of influence on school climate were recognised; local
56
beliefs, attitudes and values; organisational characteristics including physical
environment, formal organisation and personal relationships; and characteristics of
groups and individuals, including socioeconomic status, racial makeup, job
performance and parent and community satisfaction and support. Climate was seen
as the mediating variable between the input and outcomes of schooling (Lindelow et
al., 1989), and measured in terms of student satisfaction and productivity. However,
the relationship between these elements was reciprocal, in that school climate both
influences and is shaped by these inputs and outcomes.
Lindelow (et al., 1989) explained the three assumptions that the model is
based on. Firstly, the quality of the school environment needs to be seen in a
longitudinal context, as traditional values and habits are difficult to change.
Secondly, a consensus between the three stakeholders (students, staff and
community) concerning what is and what is not important in the educational process
is an important indicator of a healthy climate. Lastly, students must be the primary
concern of the school.
Profile of Organisational Characteristics (POC) Measure
Likert and Likert (1976) posited a conceptualisation of school climate that
“was based on a theory concerned with the relationship of superordinates and
subordinates” (Miskel & Ogawa, 1988, p. 292). Likert et al. (1976) identified eight
organisational variables that could be used to profile an organisation. They included:
1. leadership processes—the extent to which superiors displayed supportive
behaviour towards others;
2. character of motivational forces—the manner in which motives are used;
3. character of communication processes—the amount of interaction and
communication aimed at achieving the organisation’s objectives;
4. character of interaction-influence process—the amount of cooperative team-
work present;
5. character of decision-making processes—the level in the organisation where
decisions are formally made;
57
6. character of goal setting or ordering—involves the manner in which goals are
usually made;
7. character of control processes—the hierarchical levels in the organisation at
which major concerns exist with regard to the performance of the control
function; and
8. performance goals and training—the level of performance goals that
superiors within the organisation seek to have achieved (Hoy & Miskel,
1982).
Likert and Likert (1976) developed the Profile of Organisational
Characteristics (POC) based on responses to descriptions of the extent to which each
one of fifty-one statements characterised their organisation. Data collected from
trialled studies supported the internal consistency of the eight basic variables and of
the instrument as a whole. Likert et al. (1976) used a classification that described
organisations in terms of System 4 (participative) through to System 1 (bureaucratic),
including two intermediate positions: System 3 (benevolent-authoritative) and
System 2 (consultative) (Hoy & Miskel, 1982; Sergiovanni, 1991).
The Profile of Organisational Characteristics instrument was subsequently
modified to make it applicable to the organisational climates of schools. This
instrument was called the Profile of a School (POS). This profile has several
versions, and can be used with teachers, administrators and students to map
perceptions of a school’s climate. A school’s score on each of Likert et al.’s (1976)
eight characteristics was determined by “calculating the mean across corresponding
items and across respondents” (Miskel & Ogawa, 1988, p. 292). The school’s
location on the Profile of a School continuum was then established by an analysis of
the profile of the characteristics.
Hoy and Miskel (1982) referred to one published research study using the
Profile of a School measure. Researcher John Hall (in Hoy & Miskel, 1982)
compared Likert and Likert’s (1976) Profile of a School with Halpin and Croft’s
(1963) Organisational Climate Description Questionnaire. Results indicated that the
two instruments were similar, having a positive correlation of .59. This indicated that
58
the more open the school climate (Organisational Climate Description
Questionnaire), the more participative the system (Profile of a School).
Likert and Likert’s (1976) Profile of a School instrument has been used in
increasingly more research studies, and results suggest that increased degrees of
effectiveness and excellence in educational settings are associated with participative
systems (Miskel & Owaga, 1988). Staff and students are most satisfied with
participatory schools and least satisfied with bureaucratic and authoritarian ones.
Summary
Researchers have identified the importance of a school’s learning
environment as a correlate of organisational outcomes such as teacher’s
contentedness and enjoyment of their work (teacher job satisfaction) and student
achievement. While several useful measures developed since the 1960s have sought
to examine the multiple facets of learning environment, rarely have their
psychometric properties been rigorously evaluated and reported.
The Relation between School Learning Environment, School
Culture and School Effectiveness
The relation between school climate and school effectiveness is well
documented in the literature. School effectiveness, according to Kelley (1989) can be
defined as the degree to which students attain the outcomes that are measures of
goals which the school and community have established as important outcomes of
schooling.
Studies conducted by Brookover, Schweitzer, Schneider, Beady, Flood and
Wisenbaker (1978) have suggested that school learning environment has a direct
impact on student achievement. Other studies (Rutter in Anderson, 1982; Lindelow
et al., 1989) found that students were more likely to show good results and scholastic
achievement by attending some schools and not others. Rutter (in Anderson, 1982)
suggested that differences in school climate or learning environment contributed to
59
differences in school performance. Purkey and Smith (1983) agreed, and in a
synthesis of the then literature stated that a student’s chance for success in learning
cognitive skills is heavily influenced by the climate or learning environment of a
school. Further, they noted that effective schools have climates that emphasise and
promote academic excellence, and have an orderly environment with high
expectation for success. Purkey and Smith (1983) were so convinced of the positive
influence of the learning environment that they proposed a theory of school
improvement based primarily on changing the climate of a school so as to promote
planning, collegial work and a school atmosphere conducive to experimentation and
evaluation.
However, it should be noted that the relation between school learning
environment and school effectiveness is a highly complex one. Sergiovanni (1991)
argued that school improvement and enhanced school effectiveness will not likely be
accomplished on a sustained basis without the presence of a favourable school
climate, yet a favourable school climate alone cannot bring about enhanced school
effectiveness. According to Sergiovanni (1991), favourable school learning
environments result in more or less effective schooling depending on the quality of
effective leadership that exists to channel climate energy in the right direction.
Favourable school learning environments combined with quality educational
leadership are key to sustained school improvement and enhanced school
effectiveness. In other words, while a favourable learning environment does not
guarantee school effectiveness, it is a necessary ingredient for it to occur.
Many writers have stressed that effective schools need effective leaders who
must maintain and develop organisational culture that is aimed at achievement of
goals and is perceived in a positive manner by both teachers and students (Hall,
1987; Keefe, Kelley & Miller, 1985; Kelley, 1989; Sergiovanni, 1991; Sweeney,
1982). To maintain and develop a school culture that is supportive of school
effectiveness, the principal should carefully monitor the learning environment of the
school and should plan for school improvement based on assessment of the school
learning environment (Kelley, 1989). Further, Kelley (1989) argued that to improve
the learning environment of a school required changes in the culture of the school, or
that staff and students need to change their perceptions about the culture of the
60
school. Whether the principal alone should carry the responsibility for creating an
effective school or healthy climate is open to debate. The principal is indeed subject
to the norms and other socialising forces of the school, but as Kelley (1989) noted,
the principal is most responsible for the learning environment of a school and the
productivity and satisfaction attained by staff and students. Kelley (1989) concluded
that the principal’s major role in exercising leadership for improvement is to provide
staff with the information, the expectations, the support and the supervision so that
the staff are able to serve as mediators and transmitters of the principal’s
expectations. Nevertheless, as Lindelow et al. (1989) noted, a principal cannot bring
about change in the norms of a school by themselves. Principals must enlist the
support of others, both inside and outside the school, if there is to be meaningful
change in the school’s climate.
Summary
The relation between school learning environment and school effectiveness
has been well established, with some researchers arguing that the differences in
school learning environments between schools are in part responsible for differences
in student performance. However, while some researchers have argued that enhanced
school effectiveness will not occur unless a favourable learning environment
coexists, there is a danger in believing that a favourable learning environment is the
only prerequisite. An important key to continuous school effectiveness lies in a focus
on quality educational leadership aimed at vigilantly monitoring learning
environment.
School Learning Environment Constructs Used in this Thesis
Australian researcher Barry Fraser and others (Fraser, 1986; Docker, Fraser
& Fisher, 1989; Dorman, Fraser & McRobbie, 1995; Fisher & Fraser, 1990; Fisher &
Fraser, 1991; Fraser, Anderson & Walberg, 1982; Fraser & Rentoul, 1982; Fraser
1983; Rentoul & Fraser, 1983; Young & Fraser, 1992;) have collaborated in the
development of a school-level learning environment instrument that aims to evaluate
teachers’ perceptions of the psychosocial climate of a school. Although
61
commonalties exist between school-level environments and classroom-level learning
environments, these two levels are distinct and can be differentiated. School-level
environments are more global. A classroom climate may be limited to involve
relationships between student and student and between student and teacher within the
boundaries of an individual room. A school based climate measure would involve
teachers’ relationships with other teachers, the head of the department and the
principal (Rentoul & Fraser, 1983).
Rentoul and Fraser’s (1983) work has its genesis in the research of Moos
(cited in Rentoul & Fraser, 1983), who delineated three conceptual approaches to
measuring human environments. These are referred to as the dimensions of:
organisational structure; personal characteristics of milieu inhabitants; and
psychosocial characteristics and organisational climate. All three dimensions are
acknowledged in the development of the Rentoul et al. (1983) School Level
Environment Questionnaire (SLEQ) approach to measuring school-level climate.
Using the dimensions of the organisational structure approach, behaviour within an
environment is influenced by structural dimensions as diverse as school size, staffing
ratios, staff salaries and number of library books. Empirical studies and literature
reviews supported the notion that the key structural dimensions in any bureaucratic
organisation are the dimensions of formalisation (the extent of rules and regulations)
and centralisation (the extent of teacher participation in decision making; Rentoul &
Fraser, 1983). Two other dimensions include complexity (the degree of
specialisation) and professional latitude (the degree of freedom allowed to teachers).
Under a personal-characteristics-of-milieu-inhabitants approach, the character
of an environment is assumed to be dependent on the nature of its members. Indices
of this approach include measures of the intelligence level of the student body,
personal orientation of the student body (academic, social, sporting), median number
of hours spent in different activities, and student membership of organisations and
clubs. Finally, the psychosocial characteristics and organisational climate approach
involves both psychological and social dimensions of the organisation as perceived
by either an insider or an outsider, within a framework of person-milieu interaction
(Rentoul & Fraser, 1983).
62
The School Level Environment Questionnaire (SLEQ) is a 56-item
questionnaire that focuses primarily on the psychosocial characteristics and
organisational climate approach to measuring environmental climate (Fraser, 1999).
The three dimensions included are:
1. Relationship Dimensions—which are used to assess the extent to which
people are involved in the environment and the extent to which they support
each other;
2. Personal Development Dimensions—which are used to assess the basic
directions along which personal growth tends to occur; and
3. System Maintenance and System Change Dimensions—which are used to
assess the extent to which the environment is orderly, clear in expectation and
responsive to change (Rentoul & Fraser, 1983).
Moos (Rentoul & Fraser, 1983, p. 24) argued that these dimensions must be
assessed in order to “provide an adequate and reasonably complete picture of any
environment”. Rentoul et al.’s (1983) instrument addressed Moos’ three categories
by using the following eight scales to measure teacher perceptions of their
psychosocial environment. These included:
1. Relationship Dimensions
• Affiliation—teachers can obtain assistance and advice and feel accepted by
colleagues;
• Student Supportiveness—there is good rapport between teachers and
students, and students behave in a responsible manner;
2. Personal Development Dimensions
• Professional Interest—teachers discuss professional matters and show interest
in further professional development;
• Achievement Orientation—teachers expect and value high student
achievement, and strong competition is encouraged between students; and
3. System Maintenance and System Change Dimensions
63
• Formalisation—teachers are expected to comply with set rules and
guidelines, and are supervised to ensure compliance;
• Centralisation—decisions are made by an individual or by a small group of
people;
• Innovativeness—the school is in favour of planned change and
experimentation, and fosters classroom openness and individualisation;
• Resource Adequacy—support personnel, facilities and resources are suitable
and adequate.
Rentoul and Fraser (1983) reported the alpha coefficient for the eight
different scales ranged from .70 to .91 (mean .82) in its validation sample, and
between .68 to .91 (mean .79) for the cross validation sample. These values indicated
that the individual School Level Environment Questionnaire scales displayed
satisfactory internal consistency. Hence, given this reliability, this instrument was
utilised in the present investigation (see Appendix 5.1).
School Learning Environment Research
There have been many reported studies in the area of school climate.
However, Miskel and Ogawa (1988) argued that the four predominant
conceptualisations of school learning environment or school climate, including those
described earlier, primarily focus on the social-system dimension (Tagiuri in
Anderson, 1982), which emphasises the relationships between persons and groups.
Therefore, these conceptualisations have led to the development of perception-based
responses to climate. Further, all have employed what James and Jones (in Miskel &
Ogawa 1988, p. 294) categorise as “perceptual-measurement organisational
attribute” approaches to the measurement of climate. Miskel and Owaga (1988, p.
294) further reported that the other “three dimensions of climate (ecology, milieu and
culture) and two methods of measuring climate (multiple measurement-
organisational attribute and perceptual measurement-individual attribute) ... have
virtually been ignored”.
64
Anderson (1982) reported the results of a variety of studies undertaken up
until the early 1980s. These results are summarised below, and follow the Tagiuri (in
Anderson 1982; in Miskel & Owaga, 1988) taxonomy:
1. ecology variables—these have generally shown low or inconsistent relationships
with student outcomes. Ecology variables, such as school appearance or school
size that are used in school climate research have been used as independent
variables in the equation. Anderson (1982) noted that ecological variables do not
operate directly on student outcomes, but rather operate as a mediating effect of
school climate;
2. milieu—these have incorporated the characteristics of groups or persons within
the school environment, and are included as school climate variables. Findings
can be described as often divergent, perhaps due to poor causal modelling.
Teacher characteristics have shown little impact on student outcomes. However,
teacher morale, which is less easy to measure, has been associated positively in a
number of studies of school climate. Student morale was also found to be
positively correlated to achievement and student self-concept.
3. social system variables—Anderson (1982, p. 399) stated that “findings in this
area are conflicting and comparisons difficult because of the diversity of
constructs measured and the differences in how they are operationalised”.
Administrative organisations have had no significant effect on student outcomes.
However, it was reported that administrative organisational structure did effect
teacher performance, thereby indirectly influencing student performance.
Anderson (1982, p. 400) reported “the relationship of the administration with
teachers appears to be particularly important in creating a climate for
achievement”. Other social system variables shown to act as important influences
of school climate included participatory decision-making, good communications
and good teacher-teacher relations. Also mentioned were teacher-student
relationships, student-shared decision-making and strong administrative
leadership in instruction.
65
4. culture variables—values and belief systems have shown consistent and definite
relationships with school climate and student outcomes, although Anderson
(1982, p. 402) points out “we still know little about how they react with other
variables”. There were several cultural variables that research had shown to be
important, including teacher commitment, peer norms, emphasis on academics,
consistency and clear goals.
Brady (1988) noted the important social systems effect the principal has in
determining the efficiency and morale of staff, and hence a school’s learning
environment. He stated his belief that school learning environment is heavily
dependent upon the leadership of the school. Where the principal makes most
decisions alone and is aloof from active group participation, school climates tend to
be closed. De Roche (1985, p. 42) agreed, and pointed out “the school principal is the
manager, organiser, coordinator and evaluator ... and what he or she does contributes
to a large degree to whether there is a positive or negative climate in the school”.
Research during the 1990s in the area of school effectiveness has re-
emphasised the important impact a positive school learning environment has in the
overall success of schools (Sergiovanni, 1991). Effective schools display clear
organisational personality, stated missions, goals and standards of performance. They
have a sense of order, purpose and direction. Teachers provide a standard of
consistency in which a work centred environment exists, and display a high optimism
and expectations that students will learn. Teachers and principals commit themselves
to breaking down barriers to instructional learning and take a positive approach to
discipline, and model behaviours they say are important (Sergiovanni, 1991).
School Level Environment Questionnaire (SLEQ) Research
Rentoul and Fraser (1983) reported that the School Level Environment
Questionnaire (SLEQ) has been trialled in a number of studies, including one aimed
at assessing change in beginning teachers’ attitudes towards individualised teaching
approaches. Data was originally collected from a sample of 83 teachers in 19
coeducational government schools (7 primary and 12 secondary). Refinements in
item construction meant that a second sample of 34 teachers each in different schools
66
could be cross-validated with the original data. A multiple regression analysis
approach was employed to explore the relationship between five of the eight of the
School Level Environment Questionnaire scales, and changes to beginning teachers’
attitudes to classroom individualisation when some background variables are
controlled. Rentoul and Fraser (1983, p. 32) indicated that one of the interesting
findings in this study was that “greater formalisation in the school environment was
significantly associated with less favourable changes in attitudes to two important
aspects of classroom individualisation, namely, independence and differentiation”.
In a second study, Rentoul and Fraser (1983) reported on a study designed to
link together school level and classroom level environments. This study examined
the relationship between teachers’ perceptions on five of the eight School Level
Environment Questionnaire (SLEQ) scales and the same teachers’ perception of the
dimension of individualisation of the classroom environment. Several statistical
techniques were applied to the study, the results providing strong support for the
existence of a sizeable and statistically significant relationship between school level
and classroom level variables. More precisely, “formalisation of the school
environment was associated with lower levels of classroom participation,
independence and differentiation” (Rentoul & Fraser, 1983, p. 33).
A third study undertaken by Fraser, Williamson and Tobin (1987) involved
the School Level Environment Questionnaire (SLEQ) instrument, which was aimed
at further examining the link between school and classroom based perceptions of
school environment among adult students in a senior college context. This study
assessed teachers’ perceptions of actual school environment and teachers’
perceptions of preferred school environment.
Rentoul and Fraser (1987) argued that, because there were adult students
rather than child or teenage students in a senior college, the scale of student
supportiveness would be better replaced by the scale of work pressure (the extent to
which the press of work dominates the job milieu). The study sample included 106
teachers from two senior colleges, who represented 80% of the total staff. Teachers
responded to both an actual and a preferred form of the school level environment
instrument. This design allowed researchers to examine the ethos of these senior
67
colleges in terms of the discrepancies between actual and preferred environments.
Researchers also had available data from previous studies, and allowed senior
colleges to be compared to conventional high school in terms of actual environment.
This sample consisted of 34 teachers from junior high school level at 34 different
government high schools in New South Wales.
Statistical analysis validated the scales used in the School Level Environment
Questionnaire (SLEQ), and included information on each of the eight scales’ internal
consistency. Alpha coefficients for each of the School Level Environment
Questionnaire (SLEQ) scales ranged between .64 and .91 on the actual form, and
between .63 and .81 on the preferred form. Mean correlations of scales with other
scales ranged between .05 and .38 for the actual form, and between .12 and .32 for
the preferred form. Rentoul and Fraser (1987, p. 227) reported that these values
“indicate satisfactory discriminant validity and suggest that the SLEQ measures
distinct, although somewhat overlapping, aspects of school environment”. The
responses obtained from the sample of 106 senior college teachers provided data
with which to evaluate the ethos of the senior schools. One way of interpreting this
data was by comparison with actual and preferred perceptions of the senior school
environments. A second way involved comparison of data obtained from the senior
colleges with data obtained from a control group consisting of 34 high school
teachers. A comparison of actual and preferred perceptions of senior college teachers
revealed significant differences between six of the eight scales. Teachers preferred
more favourable environments in the dimensions of affiliation, professional interest,
centralisation, innovativeness, resource adequacy and work pressure than the levels
of these dimensions actually present. Where achievement orientation and
formalisation were concerned, teachers perceived that the levels of these dimensions
present in their schools was sufficient already, and did not need to be increased.
Rentoul and Fraser (1987, p. 229) concluded:
by providing a separate site for self-motivated adult learners and for teachers wishing to work with these students, the senior colleges have achieved an environment in which teachers perceive a press for student achievement and an absence of rules which teachers have to follow.
In the second comparison, seven of the eight actual school environment
scales from the senior colleges were compared with the same scales from
68
conventional high schools. In the case of affiliation, conventional high schools were
perceived as significantly more favourable than the senior colleges. This, the
researchers argued, may reflect the amount of time a teacher in a senior college stays
in such a school. This limited time would therefore affect the opportunity to build up
personal relations with adult learners. Senior colleges were perceived to have
considerably more favourable environments than conventional high schools in the
dimensions of professional interest, achievement orientation and innovativeness.
This can be explained by the selective nature of choosing teachers for senior
colleges, greater motivation to achieve among mature, adult learners and an absence
of traditional structures and traditions that would act to restrict innovativeness
amongst the staff. Generally, the data provided evidence for “the success of senior
colleges in promoting a favourable school ethos” (Fraser et al., 1987, p. 229).
Young (1998) reported a longitudinal study designed to examine rural school
effectiveness using a number of measures, including a version of the School Level
Environment Questionnaire (SLEQ). The study was designed to identify the
characteristics of effective rural high schools by examining those factors that
influenced student achievement in mathematics and science. Central to this study
was the influence of learning environment, where Young (1998) argued that previous
effective school research had demonstrated that strong links had been found between
student outcomes and learning environment. As well as measures of school level and
classroom level learning environment, measurement was made of student
background, self-concept (of both teacher and student), teacher morale and
achievement scores in mathematics and science at two points of time. The study was
designed in three phases. In the first phase, survey instruments were developed and
piloted. In the second phase, a longitudinal survey was undertaken over three years in
28 public and independent Western Australian high schools. Critical during this
phase was an evaluation of the learning environments with the schools surveyed,
both at class and at school levels. A version of the School Level Environment
Questionnaire (SLEQ) was used to gather learning environment data, which
measured the following eight environmental variables: student supportiveness,
affiliation, professional interest, mission consensus, empowerment, innovation,
resource adequacy and work pressure. A case approach was employed in the third
phase of this study, whereby the learning environment of some exceptional schools
69
in both rural and urban locations in Western Australia was examined. The selection
of exceptional schools was to be based on analysis of outlier schools from data
gathered during the second phase of the study.
Statistical analysis of all the scales used in the first two phases of this study
indicated significant differences for all scales, except general self-concept, between
rural and urban locations. As far as environmental factors were concerned, rural
students expressed greater satisfaction with their schools than did students in urban
locations. Rural students felt “teachers were more supportive, friends were more
supportive, and (they) generally felt safer” than students from urban schools (Young,
1998, p. 13). Teacher morale proved to be a useful measure in differentiating
between effective and ineffective schools. “Effective schools had more positive
classroom learning environments”, especially in the areas of perceived teacher
supportiveness and clearly defined school mission (Young, 1998, p. 22). Young
however, sounded a note of warning as far as relating effective schools with student
achievement is concerned. She noted that once adjustments were made for student
background effects, variability between schools in terms of student achievement is
minimal, and that other measures, such as staff morale and teacher efficacy need to
be taken into consideration.
Summary
Recent effective schools research has underscored the importance of school
learning environment as one of the five major determinants of increased student
achievement (Butler, 1995; Johnson, Livingston, Schwartz & Slate, 2000; Lezotte,
1991). It has also been noted that principals, although not having a direct effect on
the achievement of student outcomes, are the essential element in determining the
organisational climate of a school (Griffith, 1999). Effective schools are led by
effective principals who manipulate organisational variables such as school learning
environment to meet their vision objectives.
70
Teacher Job Satisfaction
The concepts of teacher job satisfaction and teacher motivation have long
been intertwined in the literature, causing some confusion. This, Dinham and Scott
(1997, p. 362) have argued, is because “both phenomena are inextricably linked
through the influence each has on the other”. Motivation refers to the stimulus for
behaviour in a particular context, while satisfaction refers to the product or result of
behaviour within a particular context (Dinham & Scott, 1997). Conceptually, teacher
job satisfaction research has its basis in the motivational theories of Maslow (1943)
and Herzberg (1959).
The Historical Development of Teacher Job Satisfaction Theory
Maslow’s (1943) Hierarchy of Needs Model
Maslow (1943) argued that the driving force that motivates a person to join
an organisation, stay in it and work towards its goals is an internal “hierarchy of
needs” (Owens, 1998, p. 142). These needs are arranged in ascending order, so that
when lower order needs have been satisfied, higher order needs appear, and
individuals attempt to satisfy them. Maslow’s (1943) hierarchy of needs model is
based on two notions (Miskel & Ogawa, 1988), namely, that the desire for the
satisfaction of wants and needs is the driving force behind behaviour, and that needs
can be arranged in a hierarchy of prepotency. The lowest order needs are what
Maslow (1943) described as basic physiological needs, and include the necessities of
food, water and shelter. Maslow (1943) argued that in modern society, we become
enmeshed in organisational life in order to earn the money that will in turn provide us
with the means to satisfy these basic needs. When basic physiological needs have
been met, a new level of security and safety needs appears in the hierarchy of needs
(Owens, 1998). These needs have greater potency than lower order needs as
individuals seek to satisfy them. Security and safety needs can refer to the attempt
that individuals make to earn job tenure by working hard. It also can refer to working
hard to achieve success so that freedom can be assured in old age.
71
The next level of need is associated with social affiliation. Once the basic
physiological needs and security and safety needs have been satisfied, individuals
will be free to try to meet the need for social interaction and affiliation with others
(Owens, 1998). Maslow (1943) argued that once lower order needs have been
satisfied, an individual will endeavour to meet their need for belonging and approval
by others (Miskel & Ogawa, 1988). These act as powerful motivators in determining
organisational behaviour (Owens, 1998). The second highest level in the hierarchy of
needs is esteem. Once the lower order needs have been met, the individual, according
to Maslow (1943), experiences the need to be recognised and respected (Owens,
1998). Here, issues of self-image are important, as an individual seeks prestige or
status opportunities in forming a sense of satisfaction. Maslow’s (1943) highest level
in the hierarchy of needs is that of self-actualisation. Hoy and Miskel (1996, p. 101)
described the drive to attain this level in terms of “the need to be what an individual
wants to be, to achieve fulfilment of life goals, and to realise the potential of his or
her personality”. A self-actualised person is one who is inner directed, seeks self-
growth and is motivated by loyalty to values, ethics and beliefs (Owens, 1998).
It is important to note that Maslow’s (1943) theory of “hierarchy of needs” is
a general theory of human motivation, and was not developed to describe
motivational behaviours of people in work generally, or teachers in schools
specifically (Owens, 1998). Even so, the theory has been used and modified to study
motivational behaviour of teachers in educational settings. For example, Porter
(Miskel & Ogawa, 1988) modified Maslow’s (1943) original theory to include
autonomy needs that lie between esteem and self-actualisation on the needs
hierarchy. Porter (Miskel & Ogawa, 1988) developed a Needs Satisfaction
Questionnaire (NSQ), and found that self actualisation needs were of critical
importance in motivating behaviour, and that esteem, security and autonomy needs
were more often satisfied in middle management positions. Trusty and Sergiovanni
(Miskel & Ogawa, 1988) applied Porter’s (1961) findings in educational settings.
Trusty and Sergiovanni (Miskel & Ogawa, 1988) found that there were deficiencies
in the higher order needs of esteem, autonomy and self-actualisation of school
administrators. Administrators had lesser esteem deficiencies and greater self-
actualised needs when compared to teachers. Trusty and Sergiovanni concluded that
deficiencies in the area of self-esteem “represented the largest source of needs
72
deficiency for teachers” (Miskel & Ogawa, 1988, p. 280). Subsequent research in the
1970s supported Trusty and Sergiovanni’s findings.
Herzberg, Mausner and Snyderman’s (1959) Two-Factor Theory
The second theory to influence the course of teacher job satisfaction research
was based on Herzberg, Mausner and Snyderman’s (1959) Two-Factor Theory.
Motivation, according to this theory, is comprised of two separate but independent
factors, namely, motivational factors (which can lead to job satisfaction), and
maintenance factors (which must be sufficiently present so that the motivational
factors will come into play; Owens, 1998). In this approach, several work
motivational factors, called intrinsic factors of work, enhance job satisfaction. These
features include responsibility, achievement, recognition and advancement.
Alternatively, maintenance factors, sometimes called work hygiene factors, enhance
job dissatisfaction. These features include supervision, pay and interpersonal
relations with superiors and peers (Conley & Levinson, 1993; Miskel & Ogawa,
1988; Owens, 1998). Based on research findings, Herzberg et al. (1959) reasoned
that one set of factors, motivators, produces satisfaction, while another set produces
dissatisfaction. They concluded that “work satisfaction and dissatisfaction are not
opposites; rather, they are separate and distinct dimensions of work orientation”
(Miskel & Ogawa, 1988, p. 280).
Several studies have investigated this theory in educational environments.
Sergiovanni (1967) and Schmidt (1976) have replicated Herzberg et al.’s (1959)
study, finding that teachers and administrators can identify one set of factors that
relate to job satisfaction, and a different set of factors with job dissatisfaction (Miskel
& Ogawa, 1988). Miskel (1973) found that administrators had a greater tolerance for
work pressure than did teachers. Further, he found that educators who aspired to
administrative positions demonstrated a greater desire for risk. Miskel (1973)
extrapolated these findings and suggested that: “people who aspire to rise to
membership in the next higher group tend to adopt the attitudes of the people of the
next level or group before attaining the promotion or membership” (Miskel &
Ogawa, 1988, p. 281). Hackman and Oldman (Conley & Levinson, 1993) used
Herzberg’s (1959) model to study the concept of work redesign and teacher
73
satisfaction. This study was based on the relation between intrinsic motivators,
psychological states of the individual and affective work outcomes. Hackman and
Oldman (1980) confirmed the strong predictive effects of intrinsic motivators on
work outcomes, and that an individual’s “needs and values moderate the relationship
between intrinsic work features and work outcomes” (Conley & Levinson, 1993, p.
455).
Dinham and Scott (1997) reported a study designed to explore further
Herzberg et al.’s (1959) two-factor theory of teacher satisfaction. Dinham and Scott
(1997) have argued that a central feature of the conceptualisation of career
satisfaction is the role that is played by need fulfilment, linking Maslow’s (1943)
work with Herzberg et al.’s (1959) theory. Job satisfaction is a dynamic construct
that “is an indicator of the degree of need fulfilment experienced by an individual”
(Dinham & Scott, 1997, p. 363). The presence or absence of these aspects of an
individual’s work will influence the satisfaction those individuals feel as a result of
their work. Dinham and Scott (1997) surveyed 2,336 respondents (teaching and
administrative staff) over 71 schools (primary and secondary schools) in the western
Sydney region. They concluded that the same, broad pattern of satisfiers and
dissatisfiers was present in the sample. Teachers and those in administrative or
promotions positions were “most satisfied by matters intrinsic to the role of
teaching” (Dinham & Scott, 1997, p. 375). These intrinsic satisfiers included student
achievement, positive relationships with students and others, self-growth,
professional development, and feeling part of collegial, supportive environments. As
expected, the major dissatisfiers were sourced from matters extrinsic to the task of
teaching. These dissatisfiers were, for the most part, outside the control of
respondents, and included the poor perception of teachers by the community, poor
media perception of teachers and their “easy” work, the rapid rate of change and the
lack of support provided by the system. Dinham and Scott (1997) found support for
the general notion of a two-factor theory of satisfaction. The major dissatisfiers were
largely external to respondents, and distracted “from the facilitation of student
achievement and teacher effectiveness” (Dinham & Scott, 1997, p. 375). The major
satisfiers were intrinsic to respondents, and assisted in the primary focus of teaching
students. However, an interesting “third domain” was revealed by the study. Falling
between the intrinsic satisfiers and extrinsic dissatisfiers:
74
are school based factors, such as school leadership, climate and decision making, school reputation, and school infrastructure … where most variation occurred from school to school and where there is thus the greatest potential for change within schools (Dinham & Scott, 1997, p. 376).
Dinham and Scott (1997) argued that this has important implications for addressing
issues of teacher job satisfaction. Attention must be given to each of the three domain
areas, given a school’s particular contextual situation, if teacher job satisfaction is to
be raised. While schools have little control over extrinsic dissatisfiers, they do
exercise a “degree of control over school based matters, such as leadership and
decision making” (Dinham & Scott, 1997, p. 376-377). Teachers themselves exercise
the greatest control over intrinsic satisfiers, including their own teaching and
professional development.
Teacher Job Satisfaction Constructs Used in this Thesis
While the primary focus of this present research is to examine the leadership
effect of principals on school learning environments, aspects of teacher job
satisfaction are considered in both the quantitative and qualitative components of this
study. Bass and Avolio (1997) have argued that transformational leadership can
produce extraordinary outcomes in terms of increased commitment to achieving
group or organisational goals. This is done through raising follower awareness to the
importance of achieving group goals, transcending self-interest for the sake of the
team, and developing followers’ needs to higher levels in areas such as achievement.
The result is that “transformational leadership encourages others to both
develop and perform beyond standard expectations” (Bass & Avolio, 1997, p. 18).
Specifically, transformational leadership behaviours account for unique variations in
performance ratings such as satisfaction, extra effort and effectiveness, more than do
transactional leadership behaviours.
The Multifactor Leadership Questionnaire (MLQ-5X; Bass & Avolio, 1997)
measures three constructs of follower satisfaction with leadership behaviour. These
factors are:
75
• extra effort—reflects the extent to which associates exert effort beyond what
is expected as a consequence of leadership;
• effectiveness—relates to perceived leader effectiveness in areas such as
meeting the job related needs of followers, representing followers’ needs to
higher authorities, contributing to organisational effectiveness; and
• satisfaction—reflects how satisfied followers are with the leader’s
behaviours, as well as how satisfied in general they are with the leader. (Bass
& Avolio, 1997).
While perceptions of follower satisfaction with leadership behaviour can be
considered quantitatively by examining the correlations between these scales,
specific satisfiers and dissatisfiers of teacher job satisfaction, or the contentedness
and enjoyment a teacher expresses towards their work will be examined through
respondent interviews and qualitatively explored. In this part of the study, the
interaction between the effects of the principal’s leadership behaviour and specific
school learning environment factors can be considered on teacher job satisfaction.
Three other teacher satisfaction constructs have also been included as
measures of satisfaction. They include perceptions of teacher influence in the
administration, of decision-making processes within their respective schools, and
perceptions of teacher effectiveness in relation to meeting the educational needs of
their students. A third, composite construct was developed in the course of the study
that examined teacher perceptions of control, or the perception of the degree to
which teachers controlled their workloads.
Herzberg et al. (1959) identified several motivators which together, acted to
satisfy workers intrinsically. These motivators include responsibility, career
advancement and the nature of the work itself, and are associated with job
satisfaction (Owens, 1998). It is reasonable, therefore, to conclude that perceptions of
teacher influence, of effectiveness and of teacher control would act as intrinsic
motivators to teachers, and provide some measure of job satisfaction.
76
Summary
Recent research has verified the role that intrinsic and extrinsic factors play in
influencing teacher job satisfaction (Latham, 1998). Dinham & Scott’s (1997) study
is particularly interesting as it suggests a third domain of factors may exist that will
act to either satisfy or dissatisfy teachers. These factors, Dinham and Scott (1997)
argued, including school leadership, and school learning environment, come under
the control of school leaders, who may well orchestrate them to increase teacher job
satisfaction in schools. The present investigation examined both intrinsic, extrinsic
and “third domain” factors in the qualitative phase of the study, where the
motivational influences of the principal’s leadership behaviour and school learning
environment were considered by respondents within their school contexts.
Leadership Behaviour, School Learning Environment and Teacher
Job Satisfaction
A rich research history has developed around the concepts of school
leadership and school learning environment over the past decade or so, especially
from within the effective schools area. However, while the concepts of school
leadership and school learning environment are closely intertwined, few educational
studies have related them together (Griffith, 1999). Further, even fewer studies have
examined their combined direct and indirect effects on teachers’ perceptions of job
satisfaction. Yet, the potential yields from such studies are immense. Paradigm
developments over the past twenty or so years have given educational researchers
new platforms upon which to examine school leadership. In particular,
transformational leadership has been advocated by some researchers as a
conceptually productive perspective from which to both define leadership and to
examine its effects, especially in times of educational turbulence (Leithwood &
Jantzi, 2000). Recent studies have demonstrated the influence that transformational
style leadership behaviours can have on school reform and the development of
collaborative cultures (Leithwood & Jantzi, 1990), group problem solving
(Leithwood & Jantzi, 1991), the process of implementing innovation (Leithwood,
Jantzi, Silins & Dart, 1992), and perceptions of teacher satisfaction (Kirby, King &
Paradise, 1992). These impacts are brought about largely by transformational
77
principals’ ability to emphasise vision (Hoover, Petrosko & Schulz, 1991; Roueche,
Baker & Rose, 1989), inspirational leadership (Murray & Feitler, 1989) and
commitment (Leithwood & Jantzi, 2000), build collaborative cultures (Leithwood &
Jantzi, 1990) especially in the process of problem solving (Leithwood & Jantzi,
1991), emphasise open and clear communications (Kirby, Paradise & King, 1992;
Roueche, Baker & Rose, 1989). Clearly, many of these leadership behaviours will
have consequences for outcomes such as teacher perceptions of job satisfaction.
Given the impacts of transformational leadership style behaviours, the
influence of the transformational principal on school learning environment needs to
be considered (Sergiovanni, 1991). Again, the effective schools literature is
suggestive that manipulation of a school’s learning environment can have potentially
positive or negative effects, not only on student outcomes (Young, 1998), but also on
teacher perceptions of job satisfaction. Brady (1988) identified the relation between
principals and school learning environment, highlighting that the more open and
positive the school climate, the greater is the teacher morale. Young (1998) agreed,
arguing that effective schools have learning environments supportive of teachers,
encouraging teacher morale and collegiality. Fraser, Williamson and Tobin (1987)
demonstrated the relation between formalisation, or the rigidity in school structure
perceived by teachers, teacher perception of achievement orientation and acceptance
of innovativeness.
All of these factors are characteristics of a school’s learning environment, and
hence have the potential to impact teacher perceptions of job satisfaction. Further,
such factors seem to also be influenced either directly or indirectly by the leadership
style behaviours of the principal.
78
Theoretical Framework
The theoretical framework underpinning the research methodology employed
in this study is based upon the mediated-effects model of effective schools (Hallinger
& Heck, 1998). Figure 2.1. below indicates the relation between antecedent
variables, the principal’s leadership behaviours, intervening school and classroom
variables and teacher outcomes. This model hypothesises that leaders will indirectly
achieve their effects on school outcomes. While leadership behaviours contribute to
the outcomes desired by schools, that unique contribution will almost always be
mediated by other factors. These factors may be external to the school environment,
and could include political directives, world events and economic conditions.
Internal factors can also contribute to mediating the effects of leadership, and could
include school policies and teacher commitment (Hallinger & Heck, 1998).
Antecedent Principal Intervening School Teacher
Variables Leadership and Classroom Outcomes
Variables
Figure 2.1. General theoretical framework guiding research on leadership, school learning environment and selected teacher outcomes (adapted from Hallinger and Heck, 1998, p. 162).
In the context of this study, this model is utilised to test the impact of
different types of leadership styles (transformational, transactional and laissez-faire)
on the intervening variables of the effectiveness of the school learning environment,
and teacher outcomes such as the effectiveness of teachers and global satisfaction
with leadership. Global satisfaction with leadership is a composite outcome that
incorporates several measures of teacher satisfaction with leadership, including
teacher perception of extra effort that is a consequence of leadership behaviour and
effectiveness of the principal as a leader. Figure 2.2. below, details specifically the
conceptual framework around which this study is based, as well as detailing those
constructs measured in this study. It is acknowledged that the principal’s leadership
79
style will be influenced by antecedent factors as diverse as socioeconomic status of
the school community and age, gender and experience within school variables. These
antecedent variables, together with the principal’s leadership style behaviours, will
influence the school variable of school learning environment. Finally, all three of
these sets of variables will ultimately influence teacher outcomes, particularly those
related to job satisfaction. In this model, the variable of the principal’s role is
assumed to be both a dependent and an independent factor (Hallinger & Heck, 1998).
As a dependent variable, the principal is subject to the influence of external
antecedent factors such as socioeconomic status, or prevailing external environment
conditions such as technological change. As an independent variable, the principal is
considered to be the agent of change, influencing directly the actions and therefore
the effectiveness of teachers, the effectiveness of learning conditions within the
school, and the attainment of desirable outcomes such as teacher job satisfaction and,
indirectly, student learning outcomes. It is also acknowledged that in this model,
antecedent variables can have an important causal influence on both mediating and
outcome variables. However, it is important to note that the focus of this study is to
critically examine the relation between different leadership styles of principals and
school and classroom variables that have been identified by previous research to be
important characteristics of effective schools (i.e. the effectiveness of school learning
environment, the effectiveness of teachers, and teacher job satisfaction).
There are several advantages in adopting a mediated effects model to serve as
the basis of the research methodology employed in this study. Firstly, mediated
effects models have been demonstrated by previous research to produce a consistent
pattern of results of the positive indirect effects of the principal’s leadership style on
school effectiveness. Further, mediated effects models have provided a more
complex representation of principal effects within schools compared to simple direct
effects or a moderated effects approach. Thirdly, such models include “concrete
indications of possible means through which leadership may achieve an impact on
the school’s outcomes and effectiveness” (Hallinger & Heck, 1998, p. 168). Lastly,
such models are amenable to more complex statistical techniques, such as multilevel
modelling, which can consider the strengths of direct and indirect effects at different
levels simultaneously.
80
81
The leadership practices of the principal are examined from a
transformational and transactional leadership framework (Bass & Avolio, 1997;
Burns, 1978). Five transformational factors are measured (idealised influence
(attributed), idealised influence (behaviour), inspiration motivation, intellectual
stimulation and individualised consideration) that have been posited in previous
research as key attributes of transformational leadership. Three transactional
constructs are measured (contingent reward, management by exception active, and
management by exception passive). One non-leadership construct, laissez-faire, is
used in this study (Bass & Avolio, 1997).
Intervening school and classroom variables addressed in this study are those
school learning environment constructs posited by Rentoul and Fraser (1983), and
include student supportiveness, affiliation, professional interest, achievement
orientation, centralisation, formalisation, innovation and resource adequacy. While
not an exhaustive list of possible school learning environment constructs, these eight
provide the basis on which to measure relationship, personal development and
system maintenance and system change dimensions (Rentoul & Fraser, 1983).
Figure 2.2. Specific conceptual framework guiding research on leadership, school learning environment and selected teacher outcomes
Antecedent Variables School Level Teacher Position Teacher Experience School Size Teacher Age Teacher Gender Principal Gender Teacher Level Teacher Position Teacher Experience Teacher Time in School School Size Teacher Age Teacher Gender Principal Gender Teacher Time with Current Principal
Principal Leadership School Level Transformational Behaviours Idealised Influence (Attributed) Idealised Influence (Behaviour) Inspirational Motivation Intellectual Stimulation Individualised Consideration Transactional Behaviours Contingent Reward Management by Exception (Active) Management by Exception (Passive)
Laissez Faire Leadership
School Variables School Level School Learning Environment Student Supportiveness Affiliation Professional Interest Achievement Orientation Formalisation Centralisation Innovation Resource Adequacy
Teacher Outcomes Teacher Level Global Satisfaction with Leadership Influence Effectiveness Control
82
Four teacher outcome measures also are addressed in this study. They include
teacher perceptions of global satisfaction with leadership, teacher perceptions of
influence, teacher perceptions of effectiveness and teacher perceptions of control.
Global satisfaction with leadership, influence, effectiveness and control together
constitute measures of teacher job satisfaction. Although these teacher outcomes do not
represent an exhaustive list of all possible outcome measures, they do examine those
selected areas where the principal’s leadership behaviour and school learning
environment impact on teachers’ perception of satisfaction within their schools.
Summary
Effective schools research has continually pointed to the fact that effective
schools are led by effective principals, who recognise they can indirectly influence
student outcomes by manipulating other school variables, including school learning
environment (Johnston, Livingston, Schwartz & Slate, 2000; Latham, 1998; Lee,
Dedrick & Smith, 1991). This chapter has traced the theoretical development of the
concepts of leadership, school learning environment and teacher job satisfaction, which
have underpinned the selection of constructs utilised in the present investigation.
Chapters 3 and 4 respectively, describe the research hypotheses and research
methodology developed to examine the relations between leadership behaviours of
principals, school learning environment and teacher job satisfaction.
CHAPTER 3
AIMS, HYPOTHESES, RESEARCH QUESTIONS:
RATIONALE
Introduction
Although the concepts of school leadership, school learning environment and
teacher outcomes are closely intertwined and have been intuitively linked, few
educational studies have examined the relation between them (Griffith, 1999). This
study is designed to examine the relation between the leadership style behaviours of
school principals, aspects of school learning environment and perceptions of teacher
outcomes in secondary schools in New South Wales. Specifically, these relations are
examined from the perspective of transformational and transactional leadership
practices (Bass & Avolio, 1997; Burns, 1978), where researchers have claimed that
performance beyond what is normally expected from followers is possible (Bass,
1985a), and followers who are motivated by leaders who manipulate their
environments to meet desired organisational outcomes (Kirby, King & Paradise,
1992).
The purpose of this chapter is to present: a) the statement of the problem
under investigation; b) the aims of the study; c) the specific hypotheses and research
questions that will be tested; and d) the rationale for the hypotheses and research
questions in the context of the extant research literature. Hypotheses are posed where
previous research provides a basis for prediction of directionality. Research questions
are posed where previous research results cannot inform directionality of predictions.
The Problem
Do secondary principals display different types of leadership style behaviours
(transformational, transactional, and laissez–faire) and if so, do these different
leadership styles result in differential effects in teachers’ perceptions of school
learning environments, job satisfaction and self-perceived ability to produce
84
desirable educational outcomes? To what extent do transactional and
transformational leadership styles assist principals to foster positive teacher
perceptions of school learning environments, job satisfaction and self-perceived
ability to produce desirable educational outcomes?
Aims
This study, in the context of a sophisticated multi-method research design,
aims to:
1. Test whether Bass et al.’s (1997) differential leadership style behaviours, that
underpin his transformational leadership model, can be identified in a sample of
New South Wales (NSW) secondary schools;
2. Test the differential impact of different types of leadership style
(transformative, transactional and laissez-faire) of secondary principals on
multifaceted school learning environment constructs at both the classroom and
school level;
3. Test the differential impact of different types of leadership style
(transformative, transactional and laissez-faire) of secondary principals on
teachers’ self-perceptions of satisfaction with their principal, and teachers’ self-
perceived effectiveness to produce desirable educational outcomes;
4. Examine the relation between different types of leadership style
(transformative, transactional and laissez-faire) of secondary principals, school
learning environment constructs and antecedent variables; and
5. Identify, using qualitative data gathered from respondents, leadership
strategies that enhance and/or erode teachers’ perceptions of school learning
environment, and teacher job satisfaction.
85
Statement of the Hypotheses and Research Questions
A series of hypotheses and research questions were developed to serve as the
basis for critically analysing both the quantitative and qualitative data collected
during the course of this investigation. As this project is multi-method by design,
hypotheses and research questions pertaining to Study 1 relate to an analysis of
quantitative data collected from 497 teachers in 52 secondary schools in New South
Wales. Research questions pertaining to Study 2 correspond to Aim 5 above and
relate to an analysis of the qualitative data collected through interviews in 3 selected
secondary schools involved in Study 1. The numbering of hypotheses and research
questions corresponds to the numbering of study aims presented above.
Hypothesis 1: The Relation of Differential Leadership Behaviours on School
Learning Environment
Hypothesis 1.1. Teachers who have experienced a transformational
leadership style will display higher scores than teachers who have experienced
transactional or laissez-faire leadership styles, on measures of the effectiveness of the
school learning environment variables identified in the School Level Environment
Questionnaire (SLEQ), namely student supportiveness, teacher affiliation,
professional interest, achievement orientation, centralisation (where decision-making
occurs), innovation, resource adequacy and formalisation (of the decision-making
procedures).
Hypothesis 1.2. Teachers who have experienced a transactional leadership
style will display higher scores than teachers who have experienced a laissez-faire
leadership style on measures of the effectiveness of the school learning environment
variables identified in the School Level Environment Questionnaire (SLEQ), namely
student supportiveness, teacher affiliation, professional interest, achievement
orientation, centralisation, innovation, resource adequacy and formalisation.
86
Hypothesis 2: The Relation of Differential Leadership Behaviours and Teachers’
Satisfaction with their Leader, Teachers’ Self-Perceived Effectiveness, Influence
and Control
Hypothesis 2.1. Teachers who have experienced a transformational
leadership style will display higher scores than teachers who have experienced
transactional or laissez-faire leadership styles on measures of satisfaction with their
leader, and ratings of their effectiveness, influence and control as teachers;
Hypothesis 2.2. Teachers who have experienced a transactional leadership
style will display higher scores than teachers who have experienced a laissez-faire
leadership style on measures of satisfaction with their leader, and ratings of their
effectiveness, influence and control as teachers.
Research Question 3: What Relation do Antecedent Background Variables have
with School Leadership, School Learning Environment and Teacher Outcomes?
Research Question 3. The antecedent background variables of teacher’s
position on staff (administrative or non-administrative), the amount of time a teacher
has been teaching, the amount of time a teacher has spent in a school, the size of a
school’s student population, teacher’s age, teacher’s gender, principal’s gender and
the amount of time a teacher has spent working with their current principal, will
demonstrate differential impact in terms of teacher perceptions of leadership style,
learning environment and teacher outcomes.
Research Question 4: What are the Leadership Behaviours that Enhance or Erode
Teachers’ Perceptions of the School Learning Environment, Satisfaction,
Effectiveness, Influence and Control?
Research Question 4.1. What do teachers perceive their principal’s
leadership orientation to be?
87
Research Question 4.2. To what extent do teachers find working in their
schools satisfying, and how does a principal’s leadership style effect teacher
perceptions of satisfaction?
Research Question 4.3. What leadership behaviours and strategies can a
principal employ that can enhance or erode a teacher’s perception of their school’s
learning environment?
Rationale for Hypotheses and Research Questions
Rationale for Hypotheses 1.1 and 1.2: The Relation of Differential Leadership
Behaviours on Learning Environment
The effective schools literature has long considered context as a key
determinant of what constitutes an effective principal, asserting that “principals are
captives of their environment” (Griffith, 1999, p. 269). The promise of
transformational leadership is that transformational leaders are more able to alter
their environments than are transactional leaders (Bass & Avolio, 1997; Kirby, King
& Paradise, 1992). Bass and Avolio (1995, p. 17) have argued that “transformational
leaders learn not only how to operate more effectively within the available context,
but how to change it to make it more receptive to their own leadership orientation”.
Transformational leaders, then, do not necessarily react to environmental
circumstances; they create and change them to suit their overall goals.
In a mediated effects model (Hallinger & Heck, 1998), it is reasonable to
expect that the leadership behaviours of principals will be associated with school
learning environment factors. Hypotheses 1.1 and 1.2 are based on previous research
(see Chapter 2) which suggests that principals who display transformational
leadership styles will be more active in changing their environmental conditions than
principals who display transactional and laissez-faire leadership behaviours.
Following Bass and Avolio (1997), it is therefore predicted that the transformational
leadership behaviours of principals will have a greater effect on school learning
environments than will both transactional and laissez-faire leadership behaviours.
88
In a similar manner, Bass and Avolio (1997) have suggested that while
transactional leadership behaviours can influence organisational context, they do so
to a lesser extent than transformational leadership practices. Bass and Avolio (1997,
p. 31) state clearly that “transformational leaders learn not only how to operate more
effectively within the available context, but how to change it to make it more
receptive to their own leadership orientation”. Laissez-faire leadership behaviours
are indicated by Bass and Avolio (1997) as leadership avoidance behaviours, where
the leader makes no direct attempt to control or modify their organisational context.
Given these leadership characteristics identified by previous research, it is further
predicted that transactional leadership behaviours will have a greater influence on
school learning environment than will laissez-faire leadership behaviours.
Rationale for Hypotheses 2.1 and 2.2: The Relation of Differential Leadership
Behaviours on Teachers’ Satisfaction with their Leader, Teachers’ Self-Perceived
Effectiveness, Influence and Control
Bass and Avolio (1997) have claimed that leaders who augment transactional
leadership strategies with transformational leadership strategies will motivate their
followers to performance levels beyond that which would have been regarded as
normal. However, while transactional strategies provide for effective leadership, the
achievement of “greater amounts of extra effort, effectiveness and satisfaction is
possible from employees by augmenting transactional with transformational
leadership” (Bass et al., 1997, p. 22). Bass and Avolio (1997) have argued that
transformational leaders have the capacity to elevate followers’ motivation to greater
levels of commitment in meeting group goals. In terms of outcome measures such as
performance and satisfaction, Bass et al. (1997, p. 30) contend that “transformational
leadership would be most highly correlated with effectiveness, followed by
transactional and non-transactional styles of leadership”. Based on these theoretical
predictions it is therefore hypothesised that teachers who have experienced principals
who display transformational leadership behaviours over transactional and laissez-
faire leadership behaviours, will score higher mean responses on satisfaction and
other teacher outcome scales than principals who display only transactional or
laissez-faire behaviours. Similarly, teachers who experienced principals who display
transactional leadership behaviours over laissez-faire leadership behaviours will
89
score higher mean responses on satisfaction and other teacher outcome scales than
principals who display only laissez-faire behaviours.
Rationale for Research Question 3: What Relation do Antecedent Background
Variables have with School Leadership, School Learning Environment and
Teacher Outcomes?
The mediated effects model proposed by Hallinger and Heck (1998) (see
Chapter 2) predicts that antecedent background variables will influence the
leadership of the principal, intervening variables such as school learning environment
factors, and teacher outcomes. These antecedent variables are both external and
internal to the school, and can include socio-economic factors, international, national
and local events and even current government disposition to public school funding.
However, the question arises as to which antecedent background variables, out of all
those possibly available, will most mediate the effect of leadership and school
learning environment on teacher outcomes?
Koene, Vogelaar and Soeters (2002) examined the effects of several
antecedent variables when researching the effects of transformational leadership
within chain organisations such as retail department stores. Among the antecedent
variables considered, Koene et al. (2002) identified store size (the number of full
time employee equivalents) as an indicator of the level of social organisation and
store context, including employee tenure and store tenure, as important variables
influencing perceptions of transformational leadership. Koene et al.’s (2002) study
indicated that the leaders’ “consideration” behaviours were demonstrated to be more
effective in achieving profit outcomes in smaller stores than they were in larger
stores. Bass and Avolio (1997, p. 12) have also reported that earlier preliminary
findings in non-educational settings based on the Multifactor Leadership
Questionnaire (MLQ) have indicated that while respondent age is unrelated to
results, the gender of leaders has an influence on where those leaders are scored in
terms of their transformational and transactional behaviours.
Given the large number of possible antecedent background variables, it is
hypothesised that a selection of potential antecedent variables identified by previous
90
research will differentially influence the leadership style of school principals, school
learning environment factors and the teacher outcomes of perception of satisfaction,
effectiveness, influence and control. These variables include: teacher experience in
their current school, teacher position in their current school, career length as a
teacher, school student population size, teacher gender, teacher age, principal’s
gender and length of service with current principal.
Rationale for Research Questions 4.1, 4.2 and 4.3: What are the Leadership
Behaviours that Enhance or Erode Teachers’ Perceptions of the School Learning
Environment, Satisfaction, Effectiveness, Influence and Control?
Research Question 4.1. Educational writers (Barth, 1990; Leithwood, 1992;
Leithwood & Jantzi, 2000; Sergiovanni, 1990; Sergiovanni, 1991) have long
considered the multiple tasks involved in the principal’s role of leading and
managing a school. Among the many leadership responsibilities of the principal are
included the transformational tasks of building and implementing their school’s
vision (Bass, 1985a; Berson, Shamir, Avolio & Popper, 2001; Lashway, 2000).
Previous research has reported that effective principals are capable both of
articulating a school vision and of encouraging staff members to internalise that
vision by incorporating it into their daily activities (Abolghasemi, McCormick &
Conners, 1999). A major thrust of this study is to examine those leadership
behaviours that have the potential to both enhance and erode teachers’ perceptions of
school learning environment and selected teacher outcomes (Aim 5). It was therefore
reasonable to consider the leadership orientations of principals within the context of
their school setting, particularly in terms of how their leadership impacted upon staff
in the course of their daily tasks and routines (Research Question 4.1). Given that
previous research offers little direction as to the direct and mediated influence of
principals’ leadership style on selected teacher outcomes, a research question was
posed to explore the impact of different leadership styles on selected teacher
outcomes.
Specifically, the “people-centred” and “task centred” aspects of school
leadership have been highlighted in the research literature as making important
contributions to successful leadership in schools (Day, 2000). Densten and Sarros
91
(1997, p. 5) identified some of the relationship oriented (people-centred) aspects of
leadership as entailing delegation (ownership for decision-making being given to
followers), mentoring (one on one contact between leader and follower with effective
two way communication) and a focus on what Densten and Sarros call “specific-
needs” (attention is given to the individual needs of the follower). Given the tension
between task orientation (high order transformational leadership style) of vision
building and the people orientation (low order transformational/high order
transactional) of individualised consideration, it was also appropriate to examine the
ways in which teachers’ perceive their principal’s leadership behaviours as impacting
on their daily work.
Research Question 4.2. Teacher job satisfaction has been suggested in the
school effectiveness literature (see Chapter 2) as an important teacher outcome. A
recent National Center for Education Statistics (1997) report on job satisfaction
among United States teachers found administrative support and leadership, school
atmosphere and teacher autonomy to be associated with teacher satisfaction. The
more favourable these factors were, the higher the satisfaction scores recorded. Other
studies have reported that the principal’s leadership plays an important role in
creating an empowering environment that is both positive and motivating, and “one
that promotes self-determination and self-efficacy” (Davis & Wilson, 2000, p. 349).
A principal’s leadership style behaviour has the capacity to influence both
directly and indirectly a teacher’s perception of work satisfaction. Research Question
4.2 is designed to examine the relation between a principal’s leadership style and
teacher perceptions of satisfaction, particularly those behaviours that heighten a
teacher’s sense of satisfaction. A research question was posed as previous leadership
research in this area had not considered the differential effects of transformational,
transactional and laissez-faire leadership behaviour on teacher job satisfaction.
Research Question 4.3. The question of the mediating effect of the
principal’s leadership behaviour on teacher perceptions of school learning
environment is examined in Research Question 4.3. This question specifically
explores the mediated effects model (Hallinger & Heck, 1998) of principal effects in
more detail, by identifying those leadership behaviours that directly influence
92
teachers’ perceptions of school learning environment, either positively or negatively.
Hallinger and Heck (1998) proposed that antecedent background factors would
mediate the effect of leadership practices and school learning environment variables
on school outcomes such as teacher job satisfaction. The goal of this research
question is to identify those leadership behaviours that principals can use to help
heighten a teacher’s perception of the efficacy of their school learning environment.
A research question was posed, given much of the previous research in this area had
relied on intuitively linking leadership behaviour with school learning environment
(Griffith, 1999), and that this link had not been considered from the perspective of
the differential influence of transformational, transactional and laissez-faire
leadership behaviour.
Summary
This chapter has outlined the problem, aims, hypotheses and research
questions to be addressed in the present investigation. A rationale for the hypotheses
and research questions has been demonstrated within the context of previous theory
and research. This theory proposed that a principal’s leadership behaviour might
influence teacher outcomes through the mediating influence of school and classroom
level variables, such as school learning environment. The next chapter describes the
research methodology that was employed to address the hypotheses and research
questions posed.
93
CHAPTER 4
METHODOLOGY
Introduction
The purpose of this chapter is to provide a detailed description and
justification of the methodology used in the present investigation. Two interrelated
multi-method studies were devised to address the hypotheses and research questions
posed (see Chapter 3). Study 1 is based on quantitative data and was devised to
examine the relation between differential leadership styles, school learning
environments and selected teacher outcomes. Study 2 is based on qualitative data and
was devised to enrich the findings by elucidating those specific leadership
behaviours which act to enhance or erode teachers’ perceptions of school learning
environment and teacher satisfaction with leadership, effectiveness, influence and
control.
Firstly, the methodology is presented for Study 1. The purpose, participants,
instrumentation, procedures, and data analysis methods are described. Secondly, the
psychometric properties of the instrumentation are presented. Thirdly, the
methodology for Study 2 is presented. The purpose, theoretical background,
instrumentation, fieldwork and data analyses are described. The purpose of this
chapter is to demonstrate the strong multi-method research design devised to address
the hypotheses and research questions, utilising powerful statistical tests and
appropriate qualitative data analysis procedures to evaluate research results.
Study 1: Quantitative Methodology
Purpose
The quantitative component of the present investigation (Study 1) is a non-
experimental, correlational research design that has as its broad purpose, exploring
the impact of leadership behaviour on: School learning environment, teacher
satisfaction with leaders, and teacher effectiveness, influence and control.
94
Sample Selection
Permission to conduct this research was sought and granted by the New
South Wales Department of Education and Training (DET). Three conditions were
stipulated by the DET: that the principal must approve the methods of gathering
information in the study, that schools may withdraw at any time from the study, and
that the privacy of schools and students be protected (see Appendix 4.1).
Initially, all DET secondary schools in New South Wales were identified
according to their departmental region. Schools within each region were listed
randomly and the first three schools within each region contacted by telephone, to
ascertain their interest in this research. If a preparedness to be involved was
expressed by the principal, then a contact person was appointed, and a material pack
containing instructions and questionnaires was sent to that person (see Appendix
4.2). A follow-up telephone call with the contact person within each school was
arranged to coincide with the delivery of the materials pack. These calls were
designed to answer any questions that may have arisen and to provide instructions
regarding the administration of questionnaires.
Instructions contained within the materials pack requested that 10 to 15 staff
members complete the instruments, along with the principal, who would complete a
modified questionnaire. Teaching staff were randomly selected, across each of the
faculty areas normally contained with a secondary school, and from a range of full-
time and part-time positions. Excluded were administrative and support staff, but
included were Head Teachers, Leading Teachers, Deputy Principals, Year Leaders,
Librarians and Counsellors.
The study employed a stratified random sampling procedure (Gay, 1992;
Tuckman, 1988). Stratified sampling ensures that the sample will be selected in such
a way that identified subgroups within the population are represented in the sample
in the same proportion as they exist in the population. This procedure also allows
some control for internal validity by screening members of the population in and out
of the study based on set stratification parameters, and thereby reduces the variability
of the study.
95
Two variables that were stratified or controlled for by exclusion included the
length of service at a school by the current principal, and the length of service by
teachers at a school. Because this study examines the principal’s effects on learning
environment and on selected teacher outcomes including satisfaction with leadership,
excluded from the population were schools whose principal had served less than two
years at their current school, and teachers who had served in their current school for
less than twelve months.
Participants
The above sample selection procedure resulted in 497 teachers and 49
principals from 52 government operated secondary schools in New South Wales,
Australia volunteering to participate in the study. Slightly more females (N = 235;
51.3%) participated in the study than males (N = 200; 43.7%). This compares
favourably with New South Wales secondary school female teachers (1999) (N =
20,224; 54.55%) and male teachers (1999) (N = 16,855; 45.45%) (ABS, 2005). The
majority of respondents having more than 10 years teaching experience (N = 340;
79.2%). Approximately two thirds (N = 303; 66.1%) of the respondents were either
full-time or part-time teachers, while the other third of respondents held promotions
(Head Teacher, Deputy Principal) or non-teaching (School Councellor, Librarian)
positions.
Instrumentation
Multifactor Leadership Questionnaire (MLQ; 5X-Short). The Multifactor
Leadership Questionnaire, MLQ, (5X-Short) developed from the conceptual and
empirical studies of Bernard Bass (1985a; 1988; Bass & Avolio, 1997), and based on
James MacGregor Burns’ (1978) seminal work, was chosen as the measure of
leadership style (see Appendix 4.3). The MLQ’s constructs form the basis of a new
paradigm for “understanding both the lower and higher order effects of leadership
style” (Bass & Avolio, 1997, p. 2). It examines those factors reported as essential
leadership behaviours of exceptional leaders as well as those behaviours of more
ordinary leadership performance (Bass & Avolio, 1997). Bass et al. (1997) defined
transformational leadership as the ability to influence others towards goal attainment
96
by changing followers’ beliefs, values and needs. By contrast, transactional
leadership concerns itself with daily organisational operations, using effort exchange
for reward as the primary means of leadership (Bass & Avolio, 1997). Conceptually,
transformational leadership “augments” transactional leadership, by adding to it and
raising followers’ awareness and their motivation to work for the achievement of the
greater vision and mission of the group.
The Multifactor Leadership Questionnaire (5X-Short) includes eight
transformational and transactional leadership constructs, and one non-leadership
factor. Bass and Avolio (1997) defined transformational leadership in terms of the
following five constructs: idealised influence (attributed), idealised influence
(behaviour), inspirational motivation, intellectual stimulation and individualised
consideration. Transactional leadership is defined in terms of the following three
constructs: contingent reward, active management by exception, and passive
management by exception. Bass and Avolio (1997) also included a non-leadership
factor, Laissez-Faire leadership, in the MLQ instrument. This factor indicated an
absence of leadership, and signalled leadership avoidance behaviour. Bass et al.
(1997) suggested that in this style of leadership, decisions were often delayed or not
made, and no attempt was made to motivate followers or to recognise and satisfy
follower needs. The MLQ consists of 36 items, 4 items each measuring each of the
nine constructs outlined. In addition, the MLQ contains 9 items related to teachers’
perceptions of satisfaction with leader behaviours. Each item is measured using a 5-
point scale, with responses ranging from not at all (0), once in a while (1), sometimes
(2), fairly often (3) and frequently, if not always (4).
Principals completed a “first-person” version of the MLQ (see Appendix 4.4),
which consisted of 36 items designed to measure each of eight leadership constructs
and one non- leadership construct. Minor modifications were made to this instrument
that would more adequately reflect the Australian environment, including changing
the term “Headmaster” to “Principal”. Again, each item is measured using a 5-point
scale, with responses ranging from not at all (0), once in a while (1), sometimes (2),
fairly often (3) and frequently, if not always (4).
97
School Level Environment Questionnaire (SLEQ). Fisher and Fraser (1990)
developed the School Level Environment Questionnaire (SLEQ) to examine the
psychosocial learning environment in schools, and to provide opportunity for
directed reflection in the application of environmental change. Three broad
categories of dimension are used to describe organisational climate—Relationship,
Personal Development, and System. The SLEQ consists of 56 items that assess eight
constructs, each measured by seven items (see Appendix 4.3). The relationship
constructs were defined as student supportiveness and affiliation. Personal
development constructs included professional interest and achievement orientation,
while the systems constructs were defined as formalisation, centralisation,
innovativeness and resource adequacy. All items are measured on a 5-point scale,
with responses ranging from: strongly agree (5), agree (4), not sure (3), disagree (2)
and strongly disagree (1). In addition, several items were negatively worded, and
these items were reverse scored.
Teacher Outcomes. Included in the MLQ-5X is a brief, nine-item survey
designed to question a respondent’s attitude towards three aspects of satisfaction with
their work, namely: extra effort, effectiveness and satisfaction (see Appendix 4.3). In
addition, six items describing two other researcher-devised constructs (based on
Patterns of Adaptive Learning Survey (Midgley, Maehr, Hicks, Roeser, Urban,
Anderman & Kaplan, 1996) were also included in the survey. These constructs
included perceptions of teacher influence and perceptions of teacher effectiveness. A
fourth construct, perceptions of teacher control, was developed from items selected
from the SLEQ (see Appendix 4.3). These items are measured on a 5-point scale,
with responses ranging from: not at all (0), once in a while (1), sometimes (2), fairly
often (3) and frequently, if not always (4).
Table 4.1 summarises the scales used in the Multifactor Leadership
Questionnaire, MLQ-5X (Short), School Level Environment Questionnaire (SLEQ)
and Teacher Satisfaction Questionnaire. An example of items in each of the scales is
provided, along with the number of items in each scale.
98
Table 4.1 Summary of Instruments—Multifactor Leadership Questionnaire, MLQ-5X (Short), School Level Environment Questionnaire (SLEQ) and Teacher Satisfaction Questionnaire
Instrument Scale Number of Items
Example of Item
MLQ-5X (Short)
Idealised Influence (Behaviour)
6 Talks about their most important values and beliefs regarding education.
Idealised Influence (Attributed)
6 Instils pride in me for being associated with him/her.
Inspirational Motivation 6 Talks optimistically about the future. Intellectual Stimulation 6 Re-examines critical assumptions to
question whether they are appropriate. Individualised
Consideration 6 Spends time teaching and coaching.
Contingent Reward 6 Provides me with assistance in exchange for my efforts.
Management by Exception (Active)
6 Focuses attention on irregularities, mistakes, exceptions, and deviations from standards.
Management by Exception (Passive)
6 Fails to interfere until problems become serious.
Laissez-Faire 6 Avoids getting involved when important issues arise.
SLEQ Student Supportiveness 7 Most students are pleasant and friendly to teachers.
Affiliation 7 I feel accepted by other teachers. Professional Interest 7 Teachers frequently discuss teaching
methods and strategies with each other. Formalisation 7 It is considered very important that I closely
follow syllabuses and lesson plans. Centralisation 7 Teachers are frequently asked to participate
in decisions concerning administrative policies and procedures.
Innovation 7 Teachers are encouraged to be innovative in this school.
Resource Adequacy 7 Video equipment, tapes and films are readily available and accessible.
Achievement Orientation 7 Teachers have to work long hours to complete their work.
Teacher Satisfaction
Satisfaction (with leader)
2 Uses methods of leadership that are satisfying.
Effectiveness (of leader)
4 Is effective in meeting organisational requirements.
Extra Effort 3 Gets me to do more than I expected to do. Perceptions of influence 3 In this school, some teachers have more
influence than other teachers. Perceptions of
effectiveness 3 I can deal with almost any learning
problem. Perceptions of control 3 I have to refer even small matters to a senior
member of staff for a final answer.
99
Demographic Survey. A brief demographic survey was included for both
staff and principals. Eight items were included in the teachers’ and principals’
survey, such as current position in school, total years of teaching experience, years at
current school, number of students at current school, age, gender and years of service
with the current principal (see Appendix 4.3). These items were prominent in the
literature as possible sources of mediated influence that could account for variations
in teacher perceptions of leadership and school learning environment, and hence
were included.
Procedures
Data was collected using the four survey instruments (MLQ-5X Short,
SLEQ, the teacher outcomes measures, and a brief demographic survey). These
instruments were combined into a single instrument booklet that respondents were
asked to complete (see Appendix 4.3). The instrument was self-administered and
required no direct intervention by either the school contact person or the researcher.
Further, principals were asked to complete the modified version of the MLQ-5X,
along with a brief demographic survey (see Appendix 4.4). The possibility of a
sampling bias was recognised at this point. It is possible that instructions contained
in the materials pack with regards to staff selection may have been ignored.
Sampling error may have occurred if principals directed staff to fill in the
instruments, or if no random assignment of staff was made. However, follow up
telephone calls to each school’s contact person reinforced the need for staff to be
randomly invited to participate in this survey. Completed instruments were collected
by the contact person, and returned to the researcher in provided envelopes.
Overview of Methodology Employed to Test Research Hypotheses Proposed
Hypotheses 1.1 and 1.2: The relation of differential leadership behaviours
on school learning environment. Hypothesis 1.1 predicted that teachers who have
experienced a transformational leadership style will display higher scores compared
to teachers who have experienced transactional or laissez-faire leadership styles for
the effectiveness of the school learning environment in supporting: Students, teacher
affiliation, professional interest, achievement orientation, centralisation (where
100
decision-making occurs), innovation, resource adequacy and formalisation (of the
decision-making procedures). Hypothesis 1.2 predicted that teachers who have
experienced a transactional leadership style will display higher scores compared to
teachers who have experienced a laissez-faire leadership style for the effectiveness of
the school environment in supporting: students, teacher affiliation, professional
interest, achievement orientation, centralisation, innovation, resource adequacy and
formalisation. In order to test these hypotheses multilevel modelling techniques were
used to explore any relations that may exist between transformational, transactional
and laissez-faire leadership style behaviours and the school learning environment
variables. Multilevel modelling allows for an examination of the relation between
sets of variables at different levels of analysis. A rationale and discussion of the
techniques involved in constructing multilevel models is found in the section below,
on Data Analysis.
Hypotheses 2.1 and 2.2: The relation of differential leadership behaviours
on teachers’ satisfaction with their leader, teachers’ self-perceived effectiveness,
influence and control. Hypothesis 2.1 predicted that teachers who have experienced
a transformational leadership style will display higher scores compared to teachers
who have experienced transactional or laissez-faire leadership styles for satisfaction
with their leader, and ratings of their effectiveness, influence and control as teachers.
Hypothesis 2.2 predicted that teachers who have experienced a transactional
leadership style will display higher scores compared to teachers who have
experienced a laissez-faire leadership style for satisfaction with their leader, and
ratings of their effectiveness, influence and control as teachers. Multilevel modelling
techniques (see discussion below) are used to examine and explore any relations that
may exist between transformational, transactional and laissez-faire leadership styles
of principals and teacher outcome measures.
Research Question 3: What Relation do Antecedent Background Variables
have with School Leadership, School Learning Environment and Teacher
Outcomes? Hypothesis 3 predicted the antecedent background variables of teacher’s
position on staff (administrative or non-administrative), the amount of time a teacher
has been teaching, the amount of time a teacher has spent in a school, the size of a
school’s student population, teacher’s age, teacher’s gender, principal’s gender and
101
the amount of time a teacher has spent working with their current principal, will
demonstrate differential impact in terms of teacher perceptions of leadership style,
learning environment and teacher outcomes. Multilevel modelling techniques (see
discussion below) are used to assess the relative strengths of relations that may be
evident between antecedent background variables and principal’s leadership style,
school learning environment variables and teacher outcome variables.
Data Analysis Procedures
Multilevel modelling. Multilevel modelling techniques can be used to further
examine the relation between a set of variables. Multilevel analysis is a statistical
modelling procedure that allows for the nested nature of data to be taken into
consideration when examining the relations between independent and dependent
variables. “Nested” data describes individuals who are grouped into larger or higher
order units (du Toit, du Toit & Cudeck, 1999). Typically, schools fall into this type
of structure, as students are taught within classes, and classes are nested within
schools (Hill & Rowe, 1996). Schools are also nested within districts or regions.
Once the groupings have been established, they tend to become differentiated from
each other. This differentiation implies that the group and its members will influence
and be influenced by membership of that group. By analysing data within this nested
structure, important group effects can be considered (Goldstein, 1995).
There are several problems associated with the analysis of nested data that
impact on the validity of statistical conclusions. These include aggregation bias,
undetected heterogeneity or regression, mis-estimation of parameter estimates and
their standard errors, and problems related to model misspecification due to lack of
independence between measurements at different levels (Goldstein, 1995). Rowe
(2000a) has argued that a failure to properly account for the nested or hierarchical
nature of data will increase the probability of Type 1 errors, which relate to falsely
rejecting the null hypothesis.
Multilevel modelling allows researchers to deal with the problems associated
with nested data. Firstly, multilevel analysis allows the effects of observed relations
among variables to be partitioned into within- and between-school components.
102
Secondly, multilevel analysis incorporates the unique effects of individual schools
into the statistical analysis. Thirdly, multilevel modelling allows separate regression
equations to be generated for individual schools, and allows model variation among
schools as multivariate outcomes are explained by school level factors (Rowe,
2000a).
While multilevel analysis allows for the development of models that account
for the nested nature of data at different levels simultaneously, it does not provide
information about interdependent effects among constructs (Goldstein, 1995).
However, these effects can be accounted for using structural equation modelling. A
fitted structural equation model can provide a means of estimating the magnitude of
direct, indirect and interdependent effects among variables, while accounting for
measurement error in both the observed and the latent variables (Rowe, 2000a).
The distributional properties of the data, particularly skewness and kurtosis,
need to be considered in this analysis. It is assumed that data will be normally
distributed, and if this assumption is invalid, the interpretation of parameter estimates
and their associated standard errors will be incorrect. Rowe (2000a) has argued that
this is especially the case with variables that are significantly leptokurtic or
platykurtic, due to under or over dispersion, and places restrictions on the
explanatory power of the fitted models. Prelis 2.30 provides a summary of the
descriptive parameters of data, particularly in relation to skewness and kurtosis. If
tests for skewness and kurtosis show latent variables to be significantly non-normal,
then “raw” composite scale scores should be recomputed as normal scores using
Prelis 2.30.
Rowe (2000a; 2000b) has described the process involved in undertaking
multilevel analysis. The first step is to determine the proportion of variance in both
the response (dependent) variable and the explanatory (independent) variable that
may be due to data structure. The proportion of variance in each of the scales
(leadership, school learning environment and teacher satisfaction) due to between-
school and within-teachers differences can be determined using a two level variance
components model fitted with an iterative generalised least squares estimation
method. This model allows a baseline from which to compare more complex models
103
that are developed from the data. The multilevel modelling computer program
MLWin (Version 1.10; Rasbash, Browne, Goldstein, Yang, Plewis, Healy,
Woodhouse, Draper, Langford & Lewis, 2000) represents the baseline variance
components model mathematically, as shown below in Figure 4.1.
Figure 4.1. Mathematical representation of baseline variance components model.
The subscript i indicates the teacher level while the subscript j refers to the school level. Subscript ij means variables vary between teachers within a school. If a variable has a subscript j only, it varies across schools but has the same value for all teachers within a school. Variables with no subscripts are constants across all teachers and schools.
The mathematical expression yij ~ N(XB,Ω) is the standard notion for a
normally distributed response variable within the fixed part of the model XB and the
variances and covariances of the random part of the model over all levels of the
model Ω. β 0 is the mean y response for teachers in all schools, and denotes the
fixed part of the model where effects are assumed to be “constant” and without
measurement error. When teacher responses have been standardised, then β 0 will
be close to zero.
The mathematical expression x0 represents the constant vector and is
typically set at 1. is a residual which will vary randomly between schools, and
will be the same for all teachers in school j (the subscript 0 indicates that it is
oju
104
attached to ). also represents the departure of the jth school’s intercept from
the overall value
x0 oju
0β . 0ije denotes a random variable with a mean of zero and
represents the sum of all other influences on . Both and ij
y oju 0ije are regarded as
the random part of the model, where random effects are not assumed to be constant,
and are measured with sampling error. It is assumed that:
1. ∼N(0, ), where is the variance of the level 2 residuals
(school level variance);
ojuu0
2σ u0
2σoju
2. 0ije ∼N(0,e0
2σ ), e0
2σ is the variance of the level 1 residuals 0ije (teacher
level variance); and
3. and oju 0ije are independent.
The variances and u0
2σ e0
2σ are referred to as random parameters and0
β is
a fixed parameter of the model. Standard errors of fixed and random parameters are
given in parentheses for all models. The likelihood ratio (-2loglikelihood) for normal
models is computed at each iteration, using an iterative generalised least squares
procedure. This statistic is used to test the significance of more complex models
(Rasbash et al., 2000).
Once the baseline variance components model has been fitted, variance
components models that contain explanatory variables can be fitted to the data.
Again, MLWin (Version 1.10; Rasbash et al., 2000) can be used to develop
multilevel variance components explanatory models for dependent variables at
different levels of analysis. Figure 4.2 below outlines the mathematical statement that
describes a two level variance components model.
105
Figure 4.2. Mathematical representation of variance components models with explanatory variable.
This model can be written as:
yij = Xij + u0j + e0ij
where Xij is a vector of explanatory variables defined at the teacher level. u0j
and e0ij are second level (school) and first level (teacher) residuals respectively (the
subscript 0 indicates that the term is attached to X0ij, which is the constant term. It is
assumed that random effects are normally distributed.
There are two usual tests of significance for explanatory variables as they are
added to the variance components model. The first test compares the ratio of the
parameter estimate with its standard error. In a large, randomly selected, normally
distributed sample (with a mean of 0 and a standard deviation of 1), the ratio of the
fixed parameter estimate with its standard error must be greater than 1.96 (the critical
t value under normal distribution) for that explanatory variable to be considered
statistically significant.
The second test of significance is the likelihood ratio test for random
parameters (the variances of u0j and e0ij). In this test, the differences between the log-
likelihoods of the explanatory model and the baseline model are compared. Following
a chi-square distribution with degrees of freedom equal to the number of new
parameters included in the model, deviations in the log-likelihood values of greater
than 7.88 are considered to be statistically significant to the p >. 05 level.
106
Levels of Analysis. Papaioannou, Marsh and Theodorakis (2003) and others
(Marsh, Kong & Hau, 1999) have warned of the inappropriateness of not considering
the group level effects of certain constructs, such as motivational climate, in their
correct contexts. This, they have argued (Papaioannou et al., 2003) makes no sense,
especially when the focus of the research is on an inherently group-level variable. In
the present investigation, the critical variables of school learning climate and
perception of leadership behaviour must be considered at both the individual level
and the group level. Clearly the level of analysis will be at the group level, because
both the school learning environment and leadership measures used in this study are
based on individual respondents’ perceptions of these group variables. In such cases,
where individual respondents have evaluated their perceptions of group constructs
like school learning environment and leadership, Papaioannou, Marsh and
Theodorakis (2003) have argued that the mean of school learning and leadership
perceptions by each member of the school group will provide a reasonable index of
these variables.
The individual response measures of school learning environment and
leadership style behaviours were aggregated and the means and factor scores
calculated using the Statistical Package for the Social Sciences (SPSS) program
(Norusis, 1993), yielding both individual level and group level data for analysis. Six
of the eight original demographic variables were also treated in the same manner,
allowing both individual level and group level analysis. The six demographic
measures treated in this manner were teacher position, teacher experience, school
size, teacher age, teacher gender and principal gender.
Study 2: Qualitative Methodology
Purpose
Study 2 was a case study approach designed to enrich and extend the findings
from Study 1 by specifically addressing a series of research questions (see Chapter
3). The purpose of an observational case study design is to “probe deeply and to
analyse intensively the multifarious phenomena that constitute the life cycle of the
107
unit” (Cohen & Manion, 1984, p. 120). Over recent years, educational researchers
have recognised the complementary advantages that have accrued to studies
employing both a qualitative and a quantitative approach (Creswell, 1994; Worthen
Sanders & Fitzpatrick, 2004). Caracelli and Greene (1997, p. 23) have argued that
mixed method studies “have the potential to produce significantly more insightful,
even dialectically transformed, understandings of the phenomenon under
investigation”. Patton (1990) identified several broad advantages that a qualitative,
case study methodology brings to a quantitative study. These include the ability to
test results obtained from a quantitative analysis, the ability to present a different
perspective, giving depth of insight that comes from the personal perspectives of
respondents, and the ability to open up issues and give insights that may have been
overlooked by a quantitative approach. It is important to note that the interpretative,
subjective methodology of the qualitative design employed in Study 2 is designed to
complement and enrich the quantitative, correlational design used in Study1.
Qualitative research methodologies refer to inductive strategies where data is
collected from people, places and discussions not readily or easily analysed using
conventional statistical methods. Their purpose is to examine the whole in its natural
context, and to gain understanding from those being observed or interviewed
(Lichtman, 2001). There are five distinctive characteristics of qualitative research.
Firstly, qualitative research uses actual or natural settings as the direct source of data,
and the researcher is the primary instrument. This means that data collected in situ
can form the basis of further questions that can be asked of subsequent informants.
Further, tape recorded interviews can be reviewed in their entirety, with the benefit
of the researcher’s insight being the key instrument for analysis. Researchers go to a
particular setting to investigate a relation or an issue within a context, because they
believe meaning separated from its context will lose significance (Bogdan & Biklen,
1998). Secondly, qualitative research is descriptive. Data is collected in the form of
words and pictures to be analysed. The data may also include tape-recorded
accounts, interview transcripts, fieldnotes, photographs, videotape, documents and
other artefacts. A presentation of the findings will contain quotations taken from the
data, which are illustrative and substantive to the results (Bogdan & Biklen, 1998).
Thirdly, qualitative research concerns itself with the process rather than with
outcomes. The main focus is on how informants interpret meaning within the context
108
of their interactions. Fourthly, qualitative analysis is inductive. Inductive analysis
starts with observation, and builds towards patterns. It is important that assumptions
are not made in advance of analysis, so that understanding can emerge from
informants’ experiences, and from within the context they occur (Patton, 1990).
Lastly, qualitative research involves understanding how people make sense from
their experiences. This implies that the informant’s perspective is important in
understanding meaning. Qualitative research involves establishing approaches and
procedures that ensure experience can be gathered from the informant’s perspective
(Bogdan & Biklen, 1998).
Theoretical Background. There are several different theoretical frameworks
available from within which to pursue a qualitative study. The focus of the study is
informed by the theoretical framework selected, which in turn provides the
perspective from which findings are interpreted and given meaning. The theoretical
perspective relevant to this current study is referred to as symbolic interactionism
(Patton, 1990). This perspective emphasises the importance of meaning and
interpretation. In it, “people create shared meaning through their interactions, and
those meanings become reality” (Patton, 1990, p. 75). Bogdan and Taylor (1998)
have suggested that this perspective encourages researchers to look for social
perspectives, their meanings and definitions.
Research Design. The design of qualitative research may be inductive or
loose, or more structured and tight. Inductive designs involve a ground theory
approach to data collection, where a conceptual framework emerges as a result of the
process of field study, and research questions become more focused as the study
develops. Meaningful settings and informants cannot be selected prior to
commencement of the study, and instruments are derived from the context of the
perspective of the informants (Miles & Huberman, 1994). While there is some merit
in this approach, a researcher needs to have some idea as to what they are looking
for, and would be wise to make the purpose of any study known prior to its
commencement (Wolcott, 1982).
Tighter qualitative designs are more appropriate for research studies dealing
with well delineated constructs (Miles & Huberman, 1994), as they provide clarity
109
and focus for the researcher from the very beginning of the study. Miles and
Huberman (1994) also suggested other reasons for favouring tighter research designs.
Firstly, a tighter design helps the researcher to be selective in the collection of data,
avoiding the trap of everything looking important. Secondly, tighter designs are more
appropriate for multicase fieldwork where many different researchers may be
collecting data in the field. A tighter research design here will help to guide the
collection of data, providing a common framework and instrumentation. Lastly,
because researchers have background knowledge, a tighter research design allows
them to see and interpret details and meaning from the complexities of the
interactions they observe. Most research design falls between these two extremes,
and a good design will allow researchers to focus on those phenomenon that are not
well known or understood (Miles & Huberman, 1994). The well-defined nature of
the constructs considered in this study lends itself to a “tighter” qualitative design
(Miles & Huberman, 1994), where a semi-structured interview technique is
employed to guide in the collection of data.
Developing a Conceptual Framework. A conceptual framework presents in
either graphical or narrative form the main factors, constructs and variables and their
presumed relations. Developing a conceptual framework allows a researcher to
delineate those variables and relations that are considered to be important and
meaningful. The conceptual framework therefore gives direction as to what data
should be collected and analysed (Miles & Huberman, 1994). The presentation and
discussion of the conceptual framework developed for this study is presented in
Chapter 6.
Sample Selection
The goal of sampling is either to produce a sample which is representative of
a chosen population, or alternatively to illuminate a situation, gain insight or collect
information about a particular event. There are two basic forms of sampling—
probability and non-probability sampling. In probability sampling, every unit in the
universe has the same possibility of being selected for inclusion in the sample, while
in non-probability sampling there is no way of estimating the chance that every unit
110
has the same probability of being selected for inclusion in the sample (Minichiello,
Aroni, Timewell & Alexander, 1995).
There are several forms that non-probability sampling may take, including
incidental sampling, consecutive or quota sampling, snowball sampling, and the form
used in this present study: theoretical sampling. In theoretical sampling, the process
of data collection is used to generate theory, and is not designed to be representative
of a particular population in a statistical sense. The value of this form of sampling is
that it allows a researcher to include variations that have been identified as relevant
to a study in terms of preliminary assumptions and provisional findings. The
researcher can therefore make decisions about what further data should be collected
in order to develop an emerging theory (Minichiello, Aroni, Timewell & Alexander,
1995). The procedure used to select the sample from which data was to be collected
is outlined in Chapter 6.
Instrumentation
While the conceptual framework, research questions and sampling plan
provide the direction that a researcher wishes to travel in, the next stage of
investigation involves obtaining the information or data. Questions will arise at this
stage about the quality of the data to be obtained, and any analysis that the data will
undergo as the researcher attempts to answer the research questions posed. There are
several types of instrumentation available, including participant observation,
interviews, fieldnotes, audiotapes, video recording and photographs. Miles and
Huberman (1994) suggested two broad approaches to questions of instrumentation.
The first approach Miles et al. (1994) suggested is one in which there is little prior
instrumentation, and where the emphasis is on construct, descriptive, interpretative
and natural validity. The alternative approach involves a large amount of prior
instrumentation, and emphasises internal validity and generalisability. Miles and
Huberman (1994) concluded that the amount and type of instrumentation should be
based on the conceptual focus, research questions and sampling criteria. An outline
of the instrument used to collect data in the present investigation is given in
Appendix 4.5.
111
Participants
Three schools out of the original 52 schools involved in Study 1 were
approached and asked to consider further involvement in this phase of the research
project. Schools were selected based on predetermined criteria, and permission
sought to enter these schools and conduct interviews. After the principal had granted
initial permission for interviews to occur, an administrative head teacher from each
of the schools approached three or four teachers, seeking their involvement. An
interview schedule with each of the teachers was arranged for each of the days of the
interviews, and included an interview with each principal. Each interview was
conducted for a period of one hour.
Overview of Interviewing Approaches
Interviewing is the process of face to face verbal interchange that occurs
when an interviewer attempts to elicit information from an informant (Minichiello,
Aroni, Timewell & Alexander, 1995). It is the means of gaining access to
information, and is done by asking questions in face to face interaction. While the
interview can take different forms, most researchers find they lie along a continuum
between structured and unstructured types (Minichiello, Aroni, Timewell &
Alexander, 1995). In structured interviews, standardised questions are asked
according to a carefully ordered and detailed interview schedule. Each informant is
asked the exact same question in the same order so that comparability can be
achieved and bias prevented between interviews. Unstructured interviews dispense
with formal interview schedules and ordering and instead rely on the social
interaction between researcher and informant to elicit any relevant information. The
unstructured interview appears as a normal everyday conversation. While control of
the direction the conversation takes is minimal, the researcher has to ensure that
relevant information is obtained. The semi-structured interview is a form of
interviewing that lies between structured and unstructured designs. The interview
schedule is developed around a list of broad ranging questions without fixed wording
or ordering, derived from theory, previous research and intuition (Grbich, 1999).
112
There are two forms the interview process can take: person to person, and
person to group. In person to group interviews, the researcher gathers a group of
informants together in order to engage them in conversation for the purpose of
research (Minichiello, Aroni, Timewell & Alexander, 1995). While there are several
different types of group interviews available for use, the following discussion is
based on person to person interviews, as this was the interview process used in the
present study.
Person to person in-depth interviews are usually conducted in a setting
familiar to the informant. There are several important assumptions underlying the
process. Firstly, they require a greater length of time for the interview to be
conducted. This is beneficial as it allows for rapport to develop between researcher
and informant. Secondly, because the interview is between researcher and informant,
issues of confidentiality and anonymity have to be considered. Thirdly, the focus of
the interview is the informant’s account, which is being sought and is highly valued.
Lastly, the researcher’s task is to retrieve the informant’s world by understanding it
from their perspective. The goal is to remove possible distorting effects that are not
part of everyday language, so that a more conversational process is appropriate
(Minichiello, Aroni, Timewell & Alexander, 1995).
An interview schedule had been developed prior to the day of the actual
interviews in each of the schools participating in this phase of the study. Informants
had been selected based on the established criteria (length of time served with the
principal, number of years of experience), and arrangements were made for the
interviews to be conducted in an area of the schools away from potential distractions.
Each interview was timed to coincide with the length of a class period in each of the
schools. In practice, interview times were between 45 and 55 minutes in length.
The interviewer’s role in the process of interviewing is central, as the quality
of the data obtained during the interview will largely depend on the quality of
relationship developed with the informant. A good interviewer has the ability to
listen intelligently, is enthusiastic and interested in what the informant has to say, is
empathic in sharing the joys and disappointments of the informant, is focused and is
able to handle contradictory information and complex situations with sensitivity
113
(Grbich, 1999). The interviewer’s intention should be to develop a sense of rapport
between themselves and the subject so that they are at ease and free to talk about
their point of view (Bogdan & Biklen, 1998). Good interview techniques include
those where the interviewer communicates a personal interest and attention to an
informant by attentive listening, appropriate facial expressions and movements such
as nodding the head. However, the interviewer’s role is more than just participating
in a conversation with an informant. Participation involves initiating the interview by
asking questions, maintaining the interview by trying to assist the informant to focus
more sharply on the issues in question, and bringing the interview to a close. Central
to the process is the function of listening. Finally, the interviewer has to balance the
function of engaging the informant in meaningful, directed conversation while
concurrently being distant enough to sustain a critical inner dialogue which enables
the analysis of data to occur (Minichiello et al., 1995).
Interviewing relies on the assumption that participants are reflective about
their experience, and are able to give a verbal account of their attitudes, beliefs and
actions. Two common methods of recording interviews are to use an audio recorder
or to take written notes, or some combination. Among the issues to consider when
making a decision as to what method of recording the interview to employ, the
researcher should consider: which method generates the most accurate data, fairness
and comfort for the informant, the validity of the data collected, and the quality of
data gathered for later analysis (Minichiello et al., 1995). Using tape recorders is a
convenient method that can be employed to gather an accurate record of an
interview. The tape recorder allows the researcher to maintain conversation in a
natural style, and so to enhance rapport as the interview progresses. Further, there is
greater analytical depth because anecdotal information is still available for the
researcher’s use (Schwartz & Jacobs, 1979). However, the tape recorder may inhibit
interaction between the researcher and the informant. Minichiello et al. (1995)
suggested that some informants may restrict their responses because the tape recorder
provides a permanent record of the interaction. Also, there may be the perception that
an informant’s perceptive can be misinterpreted at a later date when the informant is
no longer available to provide a correction. One way to give the informant an
increased sense of control during an interview is to locate the recording device within
their reach so that the stop/pause button may be accessed. This, accompanied by a
114
simple invitation to the informant to stop the interview if they feel uncomfortable, is
often enough to reassure the informant during the process of interviewing.
Each interview commenced with a brief introduction where informants were
explained the broad purpose of the discussion. Issues of anonymity were discussed,
and informants given the opportunity to read and sign a confidentiality agreement
(see Appendix 4.6). Permission was also sought to tape record the interview, so that
the researcher could more fully focus on the direction the interview was taking rather
than on taking notes. In each case, respondents were given clear instructions that the
interview could be terminated at any time, and that the audio recorder could be
switched off simply by the respondent pressing a button.
Interview protocol refers to a full list of questions that is asked of informants,
including the appropriate order of those questions and instructions on how certain
questions are asked, and probes that are designed to provide supplementary
information. There are several types of questions that can be asked. Descriptive
questions ask informants to provide a description of an experience, issue or event.
Opinion or value questions are used to gather information on what the informant felt
about an experience, issue or event. Information of a factual nature can be
ascertained by asking a knowledge question, and probing or secondary questions
allow the researcher to clarify information or gain greater detail than may have been
available from asking the primary question. Other types of questions include
structural, contrast and sensory questions. Prompts can be used to help the informant
think about the question from another perspective, and are useful when the informant
doesn’t seem to understand the primary question that has been asked (Minichiello et
al., 1995). The interview protocol developed for this study is presented in Appendix
4.5.
Several factors also need to be considered when closing an interview. It is
important that the researcher shows that informants are respected, that their stories
have been valuable and that the time spent with the informant has been appreciated.
Minichiello et al. (1995) suggested at least two reasons why care needs to be
exercised when closing the interview. Firstly, the researcher may wish to re-
interview the informant at some later stage, so it is important that even though the
115
interview is about to be terminated, rapport is maintained. Secondly, the social
process of interviewing may create some expectations that evolve as a result of the
relationship between interviewer and subject. This is particularly important regarding
fulfilling prior promises or actions made to the informant after the research is
completed.
There are several verbal and non-verbal techniques that can be used to bring
an interview to a close. These include explaining the reason for closing, using
clearing house questions which focus on areas that have not been adequately covered
up to that point of time, or summarising the interview. Simply expressing thanks and
satisfaction for the involvement of the subject is a further way of bringing an
interview to a close. Non-verbal techniques may include looking at a clock or closing
a notebook.
Overview of the Methodology Employed to Test Research Questions Proposed
Research Question 4.1 investigates teachers’ perceptions of their principal’s
leadership orientation. To examine these perceptions, teachers will be asked a series
of semi-structured questions regarding their principal’s leadership behaviours
specifically related to the creation, building and implementation of their school’s
vision and to the relationship and team building activities of their principals (see
Appendix 4.5).
Research Question 4.2 examines the extent to which teachers find working in
their schools satisfying, and how a principal’s leadership style affects teacher
perceptions of satisfaction. Teachers were asked a semi-structured question relating
to their perceptions of those factors that make working in their current schools a
satisfying experience (see Appendix 4.5).
Research Question 4.3 examines teachers’ perceptions of leadership
behaviours and the strategies a principal can employ to enhance or erode a teacher’s
perception of the school’s learning environment. To examine this question, teachers
were asked a series of semi-structured questions regarding their principal’s influence
on various school learning environment factors, including decision-making and the
116
effects of innovation, in order to determine those strategies which have the potential
to either enhance or erode teachers’ perceptions of their school’s learning
environment (see Appendix 4.5).
Data Analysis
Data analysis is the process of bringing order, structure and interpretation to a
mass of data that has been collected in the field (Marshall & Rossman, 1999). The
goal of data analysis is to work with data in such a way that it can be organised,
broken into manageable units, synthesised and searched for patterns, thereby
discovering what is important and presenting it in a form that others will understand
(Bogdan & Biklen, 1998). Bogdan and Taylor (1998) suggested a five-point
methodology useful for the analysis of data that was selected to be utilised to analyse
the results pertaining to the present investigation. Firstly, Bogdan et al. (1998)
encouraged researchers to thoroughly read and reread all fieldnotes, transcripts,
documents and other collected material. They further suggested that a notebook be
kept so that important ideas or interpretations that occur can be tracked. Secondly,
Bogdan et al. (1998) suggested that the researcher looks for emerging themes or
patterns. These could be conversation topics, vocabulary, recurring activities,
meanings, feelings or sayings that stand out in the data. Thirdly, Bogdan et al. (1998)
directed that typologies or classification schemes be constructed that assist in
identifying themes and developing concepts and theory. This, they argued,
encourages the researcher to make conceptual linkages between seemingly different
phenomena, which in turn helps in building theory. The fourth step involves
developing concepts and theoretical propositions. This process allows the researcher
to move from data description to interpretation and theory. Concepts are abstract
ideas generalised from the data, and are used to illuminate social processes and
phenomena that may not be apparent, through a description of specific instances.
Bogdan and Taylor (1998) argued that developing concepts is an intuitive process
which starts by examining words and phrases in an informant’s own vocabulary that
capture meaning. As themes develop within the data, informants’ statements can be
compared to see if concepts unite them. Propositions are general statements that are
grounded in the data. These generalisations develop by studying themes, constructing
typologies and relating different pieces of data with each other. While concepts may
117
or not fit the data, propositions are either right or wrong (Bogdan & Taylor, 1998).
The last step in the process of data analysis involves reviewing the literature. While
Bogdan et al. (1998) suggest a minimal commitment to the literature and associated a
priori assumptions when commencing a qualitative study, in the later stages of
analysis other studies can provide useful concepts and propositions that are useful in
data interpretation. However, they warn that data should not be forced into another
researcher’s framework.
Unit of analysis. Content analysis is a technique that can focus on several
different elements of an interview. These elements can include words and phrases,
and also gestures, such as hand and head movements. At a basic level, these elements
can be counted in terms of their frequency of appearance throughout an interview.
On another level, they can be examined for the meaning associated with them. In this
present study, words are the smallest units of meaning that are examined. Clusters of
words that have been grouped together into phrases are examined, along with
sentences and paragraphs, for the meaning they convey (Minichiello et al., 1995).
Content analysis is utilised in the present investigation to examine the data collected
from the respondents interviewed in the three schools involved in this phase of the
study.
Case analysis versus cross-case analysis. Patton (1990) argued that the first
decision that has to be made in the process of data analysis is whether to begin with a
case analysis or with a cross-case analysis. A case analysis requires the researcher to
write a case study for each person interviewed or for each critical event studied.
Alternatively, a cross-case analysis means the researcher has to group together
informants’ responses to common questions so that different perspectives can be
analysed on a central issue. Where a standardised open-ended interview technique is
used, Patton (1990) recommended that a cross-case analysis is a useful approach to
take, as the interview guide can be used as a descriptive analytical framework around
which answers from different informants can be grouped by topic. This was the
approach selected in the present investigation.
118
Summary
This chapter has detailed the theoretical rationale and logic underpinning the
present investigation. The study consists of a mixed method approach, where
quantitative findings from the analysis of an instrument survey are supplemented
with the qualitative findings gained from a cross-case interview approach. The
following two chapters provide a thorough discussion of the results obtained from an
analysis of the quantitative and qualitative studies respectively.
119
CHAPTER 5
STUDY 1 RESULTS:
THE NATURE AND RELATION OF PRINCIPAL
LEADERSHIP BEHAVIOURS AND SCHOOL
LEARNING ENVIRONMENTS AND TEACHER
OUTCOMES
Introduction
This chapter presents the results of a multilevel investigation into the effects
of principals’ leadership behaviours on selected aspects of school learning
environment and teacher outcomes. This chapter also outlines the relation between
the latent and explanatory factors examined in this study, at both teacher and school
levels. Results are presented in the order and context of the hypotheses and research
questions posed in Chapter 3. Firstly, results pertaining to the influence of three
leadership style behaviours on selected school learning environment variables are
discussed. Secondly, the effects of these same leadership style behaviours on four
teacher outcome measures are considered. Finally, the influence of antecedent
variables on leadership style behaviours, school learning environment constructs and
teacher outcome measures is discussed.
Development of Constructs Used in this Analysis
Appendix 5.1 outlines the development and psychometric properties of the
constructs used in each of the leadership, school learning environment and teacher
outcomes models. These models were subsequently utilised in the multilevel
modelling techniques employed to explore each of the hypotheses and research
questions pertaining to this aspect of the present investigations, presented below.
120
Results: Hypothesis 1.1 and 1.2: Relation of Transformational and
Transformational Leadership and School Learning Environment
Overview
Hypothesis 1.1 predicted that teachers who have experienced a
transformational leadership style will display higher scores compared to teachers
who have experienced transactional or laissez-faire leadership styles for the
effectiveness of the school learning environment in student supportiveness, teacher
affiliation, professional interest, achievement orientation, centralisation (where
decision-making occurs), innovation and resource adequacy.
Hypothesis 1.2 predicted that teachers who have experienced a transactional
leadership style would display higher scores compared to teachers who have
experienced a laissez-faire leadership style for the effectiveness of the school
environment in students’ supportiveness, teacher affiliation, professional interest,
achievement orientation, centralisation, innovation, resource adequacy and
formalisation. In order to test these hypotheses, a series of multilevel models were
used to compare transformational, transactional and laissez-faire leadership styles.
Analysis
A detailed analysis of the multilevel modelling technique (also see Chapter 4)
employed to examine school level and teacher level influences of student
supportiveness is presented below. While subsequent analyses follow the same
procedures, they are presented below in summary form.
Student Supportiveness. Table 5.1 shows the result of fitting multilevel
regression models to student supportiveness (SSs) scores. Student supportiveness can
be defined as teacher perceptions of the level of rapport between teachers and
students, and the perception that students behave in a responsible manner.
121
Initially, a two level variance component model was fitted to the data in order to
determine the proportion of variance in both the response and explanatory variables.
The results are shown in the mathematical model (Figure 5.1) below for the response
variable student supportiveness, and again in Table 5.1 (Multilevel Model 1).
Figure 5.1. A baseline, two level variance component model for student
supportiveness (nsst), as shown in Table 5.1 (Multilevel Model 1). In the model,
nsstij is the outcome measure for teacher i of the jth school. XB = fixed part of the
model, Ω = covariance matrix; β0ij = intercept; u0j = random school effect; e0ij =
random teacher effect.
School intercepts (β0ij) are close to zero, reflecting that they have been
standardised and then normalised. Normalisation transformation was applied because
of considerable levels of non-normality in the raw data. All data was then
standardised so that variables would have a common metric, allowing comparisons to
be made between variables. The ratio of parameter estimates to their associated
standard errors yielded significant t-values (t greater than› 1.96), indicating that at
the school level (u0j) and the teacher level (e0ij) residual variance is statistically
significant at the p < .05 level for the response variable of student supportiveness.
Table 5.1 (Multilevel Model 1) indicates that 26.8% of the recorded variation in the
student supportiveness (SSs) is due to differences between schools, and 73.2% is due
to differences within teachers. These variation percentages are obtained by firstly
adding the parameter estimates together at both the teacher and the school level, then
by placing the teachers’ parameter estimate over the combined estimates and then
multiplying by 100, and then doing the same for the school’s parameter estimate.
Further, this variation is statistically significant (p < .05) at both levels.
122
Figure 5.2 shows the ranked residuals plotted for each of the school level
student supportiveness data sets. At each end of the graph are clusters of schools
where the confidence intervals for their residuals do not overlap the zero line. These
residual intervals indicate school departures from the overall average line predicted
by the fixed parameter. Those residual intervals that fall across the zero line indicate
schools that do not differ significantly from the average line at the 5% level
(Rasbash, Browne, Goldstein, Yang, Plewis, Healy, Woodhouse, Draper, Langford &
Lewis, 2000).
Figure 5.2. Residuals of school level student supportiveness (SSs) scores, ranked for
52 schools, showing 95% confidence intervals.
Multilevel Model 2 (Table 5.1) indicates the result of fitting the three
leadership explanatory variables, vision (VIs), individualised consideration (ICs) and
laissez-faire leadership (LFs) to teachers’ student supportiveness scores. This was
done in order to assess whether the response variable was in any way influenced by
the leadership explanatory variables without first taking into account the
complicating influence of other explanatory variables. This is shown mathematically
in Figure 5.3 below. The ratio of the parameter estimate and standard error for each
of the leadership explanatory variables indicates that none of the three leadership
factors are statistically significant in influencing teachers’ perceptions of student
123
supportiveness A non-significant decrease in the log likelihood statistic also indicates
that the three leadership explanatory variables were not influential in affecting
student supportiveness scores.
The suffixes “s” and “t” were used at the end of each variable name (in
Tables, Figures and text) to indicate the level of analysis, namely, at either school (s)
or teacher (t) level. It also needs to be noted that all school level analysis data are
derived from all individual teacher data gathered at a particular school and averaged
together. This same pattern has been applied to all school level data throughout this
analysis. Furthermore, to minimise confounding the analysis for each of the school
learning environment variables, it was decided not to include the school level
complementary variable in an analysis of the teacher level perception of that same
variable.
Figure 5.3. The effect of leadership explanatory variables on teachers’ student
supportiveness (nsst) scores, as shown in Table 5.1 (Multilevel Model 2).
In the model, nsstij is the outcome measure for teacher i of the jth school. XB = fixed
part of the model, Ω = covariance matrix; β0ij = intercept; u0j = random school effect;
e0ij = random teacher effect. Explanatory leadership variables have been standardised
and shown as zvisj (vision); zicsj (individualised consideration); zlfsj (laissez-faire
leadership), at the school level.
Four “blocks” of explanatory variables were examined to account for the
variance found in this response variable. Multilevel Model 3 (Table 5.1) indicates the
results of adding in the first block of school level explanatory demographic variables.
This is shown mathematically in Figure 5.4 below.
124
Figure 5.4. The effect of school average demographic explanatory variables on
teachers’ student supportiveness (nsst) scores, as shown in Table 5.1 (Multilevel
Model 3).
In the model, nsstij is the outcome measure for teacher i of the jth school. XB = fixed
part of the model, Ω = covariance matrix; β0ij = intercept; u0j = random school effect;
e0ij = random teacher effect. Explanatory school level demographic variables are
shown as zteachsj (teacher position); ztexpsj (teacher experience); zssizsj (school
size); ztagesj (teacher age); ztgensj (teacher gender) and zpgensj (principal gender).
This block of school-average level demographic explanatory variables, together
explained 9.1% of the variance recorded in the data scores for teachers’ student
supportiveness (SSs) scores. This is determined by adding the teacher and school
level parameter estimates together from Multilevel Model 3, and then calculating the
change as a percentage in parameter estimates from those recorded in Multilevel
Model 1. An analysis of the effect of a block of school level demographic variables
indicated that two variables, teacher experience (Texps) and teacher gender (Tgens)
had a positive, statistically significant influence on teachers’ perceptions of student
supportiveness (SSs) scores. School-average teacher experience (Texps) is the
average experience level of teachers who participated in the survey, while teacher
gender (Tgens) is defined as the proportion of female or male teachers in a particular
school. These results suggest that as the school-average level of experience
increases, teachers’ perceptions of student supportiveness increases. The statistically
significant, positive influence of teacher gender (Tgens) on student supportiveness
(SSs) scores is reflective of the fact that the schools with greater proportions of
female teachers perceive student supportiveness to be higher.
125
Table 5.1
Variation in Teachers’ Student Supportiveness Scores (N = 458) in 52 Schools
Explanatory Variables
Multilevel Model 1
Multilevel Model 2
MultilevelModel 3
MultilevelModel 4
Multilevel Model 5
MultilevelModel 6
Reduced Model 6
Fixed:
Constant
(β0jX0): -.005(.081) -.002(.080) .001(.070) .001(.071) -.004(.069) -.009(.057) -.011(.060)
School Level Demographics
X1 Teachs (β1) .105(.071) .122(.073) .128(.072) -.008(.070)
X2 Texps (β2 ) .274(.109)* .236(.115)* .252(.114)* .386(.113)* .348(.096)*
X3 Ssizs (β3) .046(.069) .048(.070) .047(.078) .036(.068)
X4 Tages (β4) -.157(.106) -.101(.110) -.130(.110) -.192(.105) -.199(.094)*
X5 Tgens (β5) .171(.071)* .169(.073)* .192(.074)* .237(.071)* -.209(.063)*
X6 Pgens (β6) .067(.072) .060(.074) .093(.078) .209(.074)* .182(.065)*
Teacher Level Demographics
X7 Teacht (β7) -.095(.043)* -.095(.043)* -.093(.043)* -.089(.040)*
X8 Texpt (β8 ) .067(.071) .069(.071) .067(.071)
X9 Tcurt (β9) -.031(.053) -.033(.053) -.034(.052)
X10 Taget (β10) -.112(.065) -.112(.065) -.111(.065)
X11 Tgent (β11) -.006(.043) -.006(.043) -.007(.043)
X12TSCPt (β12) .018(.054) .014(.054) .023(.054)
School Level Leadership
X13 VIs (β13) .102(.123) -.002(.119) .060(.108)
X14 ICs (β14 ) -.133(.122) .015(.131) .065(.120)
X15 LFs (β15) .021(.108) .120(.103) .257(.112)* .185(.066)*
School Level School Learning Environment
X16 SSs(β16) na na
X17 AFFs(β17) .048(.077)
X18 PIs (β18) -.139(.091)
X19 CENs (β19) .000(.097)
X20 INNs (β20) .061(.110)
X21 RAs (β21) .266(.070)* .239(.060)*
X22 AOs (β22) -.149(.073)* -.133(.066)*
Random:
u0j (School-level) .252(.067)* .239(.064)* .161(.049)* .170(.050)* .157(.047)* .082(.032)* .100(.036)*
e0ij (Teacher-level) .689(.048)* .690(.048)* .694(.049)* .677(.047)* .677(.047)* .680(.048)* .685(.048)*
% of Variance
Explained 1.3% 9.1% 9.9% 11.4% 19.0% 16.6%
-2 (Log-
Likelihood) 1200.875 1199.389 1187.909 1179.201 1176.857 1159.288 1167.733
Note. IGLS solutions shown; fitted parameter estimates with standard errors in parentheses. * Statistically significant beyond the p < .05 level by univariate two tailed test: i.e., the
parameter estimate is greater than twice its standard error; “na” indicates that the associated
school learning environment variable was omitted from the analysis.
126
The large decrease in school level parameter estimates when compared with
teacher level parameter estimates in the random part of the model (Model 1;
.252(.067) to Model 3; .161(.049)) indicates that much of the variance at the school
level can be explained in terms of this block of school-level demographic variables.
The change in the log likelihood statistic (log likelihood at Multilevel Model 1 minus
log likelihood at Multilevel Model 3; 1200.875—1187.909) is also indicated as
statistically significant.
Multilevel Model 4 (Table 5.1) shows the influence of adding in teacher level
demographic explanatory variables in accounting for variation in teachers’ student
supportiveness scores. Figure 5.5 displays this influence mathematically. The teacher
level demographic explanatory variable of teacher position (Teacht) is indicated as
having a statistically significant, negative influence on teachers’ student
supportiveness scores. This suggests that more experienced teachers perceive that, as
teachers become more senior in terms of their administrative position in schools, they
become less supportive of students. This may well reflect a department head’s focus
on curriculum, teaching, and learning issues rather than on student welfare issues.
The school level demographic explanatory variables teacher experience
(Texps) and teacher gender (Tgens) continue to indicate statistical significance, as
was previously explained. However, there was a marginal decline in the parameter
estimates for each of these variables.
The random part of the model indicates that the variance accounted for by the
combination of these explanatory variables is statistically significant, both at the
school level and at the teacher level. The analysis shows that residual variance at
teacher level actually increases, whereas there is a decrease in the residual variance
at the school level. As expected, the introduction of this block of teacher-level
demographic variables led to a decrease in residual variance at the teacher level (.694
in Model 3 vs. .677 in Model 3; see Table 5.1). The change in the log likelihood
statistic between Multilevel Model 3 and Multilevel 4 also confirms the statistical
significance of this block of explanatory variables.
127
Figure 5.5. The cumulative effect of teacher level demographic explanatory variables
on teachers’ student supportiveness (nsst) scores, as shown in Table 5.1 (Multilevel
Model 4).
In the model, nsstij is the outcome measure for teacher i of the jth school. XB = fixed
part of the model, Ω = covariance matrix; β0ij = intercept; u0j = random school effect;
e0ij = random teacher effect. Explanatory school level demographic variables are
shown as: zteachsj (teacher position); ztexpsj (teacher experience); zssizsj (school
size); ztagesj (teacher age); ztgensj (teacher gender) and zpgensj (principal gender).
Explanatory teacher level demographic variables are shown as: zteachtij (teacher
position); ztexptij (teacher experience); ztcurtij (teacher time in current school);
ztagetij (teacher age); ztgentij (teacher gender); and ztscptij (teacher time with current
principal).
Multilevel Model 5 (Table 5.1) shows the cumulative effect of adding three
leadership explanatory variables in combination with school-average and teacher
demographic variables in accounting for variance in teachers’ student supportiveness
scores. Figure 5.6 shows this relation mathematically. Again, the explanatory
variables (teacher experience (Texps), teacher gender (Tgens) at school average
level, and teacher position (Teacht) at teacher level) that were indicated in earlier
models are still shown to be statistically significant. However, the addition of the
three leadership explanatory variables demonstrated no statistical significance. The
change in the log likelihood statistic also indicates that there is no statistically
significant influence of the three leadership variables on teachers’ student
supportiveness (SSs) scores.
128
Figure 5.6. The cumulative effect of school level leadership explanatory variables on
teachers’ student supportiveness (nsst) scores, as shown in Table 5.1 (Multilevel
Model 5).
In the model, nsstij is the outcome measure for teacher i of the jth school. XB = fixed
part of the model, Ω = covariance matrix; β0ij = intercept; u0j = random school effect;
e0ij = random teacher effect. Explanatory school level demographic variables are
shown as: zteachsj (teacher position); ztexpsj (teacher experience); zssizsj (school
size); ztagesj (teacher age); ztgensj (teacher gender) and zpgensj (principal gender).
Explanatory teacher level demographic variables are shown as: zteachtij (teacher
position); ztexptij (teacher experience); ztcurtij (teacher time in current school);
ztagetij (teacher age); ztgentij (teacher gender); and ztscptij (teacher time with current
principal). Explanatory school level leadership variables are shown as: zvisj (vision);
zicsj (individualised consideration); and zlfsj (laissez-faire leadership).
Multilevel Model 6 (Table 5.1) indicates the cumulative effect of adding in
six school-average level school learning environment explanatory variables in
accounting for variation in teachers’ student supportiveness scores. This is shown
mathematically in Figure 5.7 below. As was the case in previous models, the school-
average level demographic variables of teacher experience (Texps) and teacher
gender (Tgens), along with the teacher level demographic variable of teacher
position (Teacht) are indicated as being statistically significant. An analysis of the
block of school learning environment explanatory variables indicates that resource
adequacy (RAs) has a statistically significant, positive influence on teachers’ student
supportiveness scores, while achievement orientation (AOs) has a statistically
significant, negative influence on these scores. The addition of the school learning
environment explanatory variables has resulted in the leadership explanatory variable
129
laissez-faire leadership (LFs) being statistically significant. Hence, after controlling
for all of the school and teacher demographic variables, the effect of laissez-faire
leadership (LFs) on student supportiveness is positive.
Figure 5.7. The cumulative effect of school level school learning environment
explanatory variables on teachers’ student supportiveness (nsst) scores, as shown in
Table 5.1 (Multilevel Model 6).
In the model, nsstij is the outcome measure for teacher i of the jth school. XB = fixed
part of the model, Ω = covariance matrix; β0ij = intercept; u0j = random school effect;
e0ij = random teacher effect. Explanatory school level demographic variables are
shown as: zteachsj (teacher position); ztexpsj (teacher experience); zssizsj (school
size); ztagesj (teacher age); ztgensj (teacher gender) and zpgensj (principal gender).
Explanatory teacher level demographic variables are shown as: zteachtij (teacher
position); ztexptij (teacher experience); ztcurtij (teacher time in current school);
ztagetij (teacher age); ztgentij (teacher gender); and ztscptij (teacher time with current
principal). Explanatory school level leadership variables are shown as: zvisj (vision);
zicsj (individualised consideration); and zlfsj (laissez-faire leadership). Explanatory
school level school learning environment variables are shown as: zaffj (affiliation);
zpisj (professional interest); zcensj (centralisation); zinnsj (innovation); zrasj
(resource adequacy); and zaosj (achievement orientation).
Together, these four blocks of explanatory variables account for 19.0% of the
variance recorded in teachers’ student supportiveness scores, the majority of which is
indicated at the school level. The change in log likelihood statistic is also supportive
of the statistical significance of these explanatory variables.
130
Figure 5.8. Statistically significant explanatory variables that influence teachers’
student supportiveness (nsst) scores, as shown in Table 5.1 (Reduced Model 6).
In the model, nsstij is the outcome measure for teacher i of the jth school. XB = fixed
part of the model, Ω = covariance matrix; β0ij = intercept; u0j = random school effect;
e0ij = random teacher effect. Explanatory variables are shown as: school level ztexpsj
(teacher experience); ztagesj (teacher age); ztgensj (teacher gender) and zpgensj
(principal gender); teacher level zteachtij (teacher position); school level zlfsj (laissez-
faire leadership); zrasj (resource adequacy); and zaosj (achievement orientation).
Reduced Multilevel Model 6 (Table 5.1) indicates the aggregated effects of
all statistically significant explanatory variables considered in this analysis. Figure
5.8 shows mathematically only those explanatory variables that have remained
statistically significant at the end of this analysis. Four variables are indicated as
having a statistically significant positive influence on teachers’ perceptions of
student supportiveness. They include teacher experience (Texps) and principal
gender (Pgens) at the school-average level, and resource adequacy (RAs) at the
school-average level. Laissez-faire leadership is also indicated as having a
statistically significant, positive influence on teachers’ student supportiveness scores.
Four explanatory variables are shown as having a statistically significant, negative
influence on teachers’ student supportiveness scores. They include teacher age
(Tages) and teacher gender (Tgens) at the school-average level, and teacher position
(Teacht) at the teacher level. Achievement orientation (AOs) is also indicated as
having a statistically significant, negative effect on teacher perceptions of student
supportiveness. Together, these explanatory variables account for 16.6% of the
131
explained variance, with the change in the school level parameter estimate indicating
that this explained variance is at this level. Interestingly, both the transformational
leadership behaviour of visionary leadership and the transformational/transactional
leadership style of individualised consideration had no statistically significant
influence on teachers’ scores of student supportiveness. This is considered an
unexpected and provocative result, given the literature in this area (see Chapter 2).
Furthermore, the laissez-faire leadership style behaviour (LFs) was indicated as
having a statistically significant, positive influence on teachers’ perceptions of
student supportiveness (SSs). This indicates that as teacher perceptions of principal’s
laissez-faire style behaviour increases, teacher perceptions of student supportiveness
also increase.
There are several factors that influenced teachers’ perceptions of student
supportiveness (SSs). While it was anticipated that principal leadership would
influence teacher perceptions of student supportiveness (SSs), it was unexpected that
student supportiveness (SSs) as a school learning environment relational dimension
(see Chapter 2), would be unaffected by principals’ individualised consideration
(ICs) behaviours. Individualised consideration (ICs) behaviours emphasise the
personal interaction between principal and staff member (Hypothesis 1.2). It was
further unexpected that a principal’s visionary leadership style would not be related
to teachers’ perceptions of student supportiveness (SSs), given the emphasis that
most schools place in their published vision statements on creating a caring and
supportive environment (Hypothesis 1.1). Contrary to expectation, so far as
leadership styles are concerned, this analysis has demonstrated that it is the
principal’s laissez-faire behaviours in conjunction with other variables that influence
teachers’ perceptions of student supportiveness (SSs). This suppression effect is
common in regression and multilevel modelling, namely that one variable, laissez-
faire (LFs) leadership becomes statistically significant as other variables are
considered in the analysis. Among the other variables considered, teacher
perceptions of resource availability (RAs) and teacher experience (Texpt) were
indicated as having a strong, statistically significant positive effect on perceptions of
student supportiveness (SSs) (Research Question 3).
132
Affiliation. Table 5.2 shows the result of fitting multilevel regression models
to affiliation scores by adding four blocks of explanatory variables. Affiliation can be
defined as the perception of collegiality among staff, and the degree to which staff
perceive they can obtain assistance and advice from other staff members.
Multilevel Model 1 (Table 5.2) indicates the results of fitting a two-level
variance components model to teachers’ affiliation scores. The random part of the
model indicates that only variance at the teacher level was found to be statistically
significant. Figure 5.9 shows the ranked residuals plotted for each of the school level
affiliation data sets. No school was indicated as departing significantly from the
overall average. This is also reflected in Table 5.2, where all of the variation in
teachers’ affiliation scores is at the teacher level.
Figure 5.9. Residuals of school level affiliation (AFFs) scores, ranked for 52 schools,
showing 95% confidence intervals.
Multilevel Model 2 (Table 5.2) indicates the effect of considering the three
leadership factors: vision (VIs), individualised consideration (ICs) and laissez-faire
leadership (LFs), as explanatory variables in accounting for variations in teachers’
affiliation scores. This set of three explanatory leadership variables accounted for
2.9% of the variance, with individualised consideration (ICs) and laissez-faire (LFs)
133
Table 5.2
Variation in Teachers’ Affiliation Scores (N = 458) in 52 Schools Explanatory Variables
Multilevel Model 1
Multilevel Model 2
MultilevelModel 3
MultilevelModel 4
Multilevel Model 5
MultilevelModel 6
Reduced Model 6
Fixed:
Constant
(β0jX0): -.011(.048) -.012(.045) -.012(.045) -.012(.045) -.012(.044) -.012(.044) -.012(.044)
School Level Demographics
X1 Teachs (β1) .031(.048) .033(.049) .060(.050) .036(.056)
X2 Texps (β2 ) .024(.081) .005(.090) -.005(.089) -.056(.112)
X3 Ssizs (β3) -.068(.047) -.065(.047) -.033(.052) -.034(.054)
X4 Tages (β4) .024(.077) .056(.084) .064(.083) .102(.094)
X5 Tgens (β5) .087(.047) .088(.050) .092(.052) .050(.065)
X6 Pgens (β6) -.050(.047) -.045(.047) .000(.050) -.012(.061)
Teacher Level Demographics
X7 Teacht (β7) -.065(.048) -.059(.048) -.066(.047)
X8 Texpt (β8 ) -.026(.082) -.031(.081) -.027(.080)
X9 Tcurt (β9) .066(.058) .065(.057) .073(.057)
X10 Taget (β10) -.052(.075) -.050(.074) -.053(.073)
X11 Tgent (β11) .021(.050) .019(.049) .021(.048)
X12TSCPt (β12) .032(.050) .057(.051) .025(.054)
School Level Leadership
X13 VIs (β13) .076(.072) .085(.080) .117(.086)
X14 ICs (β14 ) .150(.070)* .152(.086) .138(.091) .182(.061)*
X15 LFs (β15) .198(.062)* .202(.067)* .248(.087)* .209(.058)*
School Level School Learning Environment
X16 SSs (β16) .000(.065)
X17 AFFs (β17) na na
X18 PIs (β18) .120(.073) .124(.048)*
X19 CENs (β19) -.030(.080)
X20 INNs (β20) -.015(.090)
X21 RAs (β21) .115(.062) .102(.046)*
X22 AOs (β22) -.021(.067)
Random:
u0j (School-level) .014(.023) .000(.000) .000(.000) .000(.000) .000(.000) .000(.000) .000(.000)
e0ij (Teacher-level) .924(.064)* .911(.060)* .920(.061)* .912(.060)* .889(.059)* .869(.057)* .887(.059)*
% of Variance
Explained 2.9% 1.9% 2.8% 5.2% 7.4% 5.4%
-2 (Log-
Likelihood) 1270.179 1257.185 1261.677 1257.805 1245.909 1235.513 1244.670
Note. IGLS solutions shown; fitted parameter estimates with standard errors in parentheses. * Statistically significant beyond the p < .05 level by univariate two tailed test: i.e., the
parameter estimate is greater than twice its standard error; “na” indicates that the associated
school learning environment variable was omitted from the analysis.
134
leadership styles indicated as being statistically significant and positive. This
indicates that as teacher perception of a principal’s individualised consideration (ICs)
and laissez-faire (LFs) leadership behaviour increases, so does teacher perception of
affiliation.
Multilevel Model 3 (Table 5.2) shows the effect of school-average level
demographic variables as explanatory variables influencing teachers’ affiliation
scores. None of the individual school level variables were indicated as being
statistically significant, and the block as a whole accounted for 1.9% of the recorded
variance, whereas all of the residual variance was at the teacher level. The change in
the log likelihood statistic between Models 1 and 3 also indicated that this block of
variables was statistically non-significant. Multilevel Model 4 (Table 5.2) adds in the
combined effect of teacher level demographic factors as explanatory variables in
accounting for changes in teachers’ affiliation scores. No individual teacher level
demographic variable was indicated as being statistically significant.
Multilevel Model 5 (Table 5.2) indicates the combined effect of the three
leadership factors as explanatory variables in accounting for the variation in teachers’
affiliation scores. While Multilevel Model 2 (Table 5.2) indicated that individualised
consideration (ICs) and laissez-faire leadership (LFs) were statistically significant,
only laissez-faire leadership (LFs) has remained statistically significant (Multilevel
Model 5) when its effect is considered in conjunction with the other school and
teacher level demographic variables.
The issue of the importance of these effects needs to be carefully considered,
as the interpretation of any results is dependent on the causal ordering of explanatory
variables as they are introduced to the model. If a variable such as individualised
consideration (ICs) comes first in the causal ordering of explanatory variables, then
the effect of individualised consideration (ICs) will be mediated by other variables as
they are considered in the analysis. Further, as this analysis consists of only a single
wave of data, it is perhaps more appropriate to consider the effect of individualised
consideration (ICs) as having a mediated effect, its influence being negated by the
inclusion of other explanatory variables.
135
Multilevel Model 6 includes the influence of school-average level school
learning environment explanatory variables. Although these additional explanatory
variables accounted for an additional 2.2% of the recorded variance (the difference in
% of variance explained between Multilevel Models 5 and 6), initially none are
indicated as having any statistical significance.
In Reduced Multilevel Model 6 (Table 5.2), all non-statistically significant
explanatory variables were removed to produce a final solution. The random part of
the model indicated that 100% of the residual variance recorded in teachers’
affiliation scores is at the teacher level. Four explanatory variables accounted for
5.4% of the recorded variation in teachers’ affiliation scores, and included
individualised consideration (ICs), laissez-faire leadership (LFs), professional
interest (PIs) and resource adequacy (RAs). Each of these four explanatory variables
was indicated as having a positive influence on teachers’ affiliation scores.
As far as the relation between school learning environment variables and
principal leadership behaviours is concerned (Hypotheses 1.1 and 1.2), this analysis
has demonstrated that teachers’ perception of affiliation within schools is influenced
by principals’ individualised consideration (ICs) and laissez-faire (LFs) leadership
behaviours. This is suggestive that as principals demonstrate increasing
individualised consideration (ICs) behaviour, teachers perceive an increase in overall
affiliation in their schools. While the relation between individualised consideration
behaviours and teacher affiliation is expected, principals’ laissez-faire leadership
style also is indicated as having a positive influence on teachers’ perception of
affiliation, and may possibly reflect a substitutionary leadership effect (Kerr &
Jermier, 1978; Swenson, 2000) that groups of teachers can experience in the
presence of disinterested leadership.
Professional Interest. Table 5.3 shows the results of fitting multilevel
regression models to teachers’ professional interest scores. Professional interest is the
degree to which teachers discuss professional matters and show interest in further
professional development.
136
Multilevel Model 1 (Table 5.3) shows the two level variance components
model for teachers’ professional interest scores. While parameter estimates (and
standard errors) indicated that variance occurs at both the school and teacher levels,
it is only residual variance at the teacher level that is statistically significant. Figure
5.10 shows the ranked residuals plotted for each of the school level professional
interest data sets. Only two schools were indicated as departing significantly from the
overall average line.
Multilevel Model 2 (Table 5.3) indicates the amount of variance that can be
attributed to the three leadership explanatory variables. Even though the change in
log likelihood statistic is statistically significant for the set of three leadership
variables, none of the effects of each leadership variable on teachers’ professional
interest scores was statistically significant. Multilevel Model 3 (Table 5.3) indicates
the influence of school-average level demographic explanatory variables on teachers’
professional interest scores. Three variables were indicated as influencing teachers’
professional interest scores. Teacher experience (Texps) and teacher gender (Tgens)
at the school-average level were demonstrated as having a statistically significant,
positive influence on professional interest scores. This indicates that staff in schools
where there is a higher proportion of experienced teachers to inexperienced teachers,
are more concerned with aspects of their professional development than in schools
where there is a higher proportion of inexperienced teachers. The influence of
teacher gender (Tgens) suggests that schools with a higher proportion of female staff
are more concerned with professional development than in schools where there is a
higher proportion of male staff. Teacher age (Tages) at the school-average level was
indicated as having a negative influence. This suggests that staff in schools where
there is a higher proportion of younger teachers are more concerned with aspects of
their professional development than in schools where there is a higher proportion of
older teachers.
Multilevel Model 4 (Table 5.3) adds the cumulative influence of teacher level
demographic explanatory variables, while Multilevel Model 5 indicates the
accumulative effect of the three leadership style behaviours as explanatory variables
in accounting for variance in teachers’ professional interest scores. No variable in
137
Table 5.3
Variation in Teachers’ Professional Interest Scores (N = 458) in 52 Schools Explanatory Variables
MultilevelModel 1
Multilevel Model 2
MultilevelModel 3
MultilevelModel 4
Multilevel Model 5
MultilevelModel 6
Reduced Model 6
Fixed:
Constant
(β0jX0): -.003(.055) -.003(.051) -.005(.047) -.005(.046) -.005(.044) -.005(.043) -.005(.044)
School Level Demographics
X1 Teachs (β1) -.024(.051) -.022(.050) -.013(.050) -.064(.055)
X2 Texps (β2 ) .217(.084)* .210(.091)* .199(.089)* .248(.100)*
X3 Ssizs (β3) -.087(.050) -.088(.048) -.064(.052) -.048(.054)
X4 Tages (β4) -.189(.081)* -.188(.085)* -.166(.084)* -.155(.088)
X5 Tgens (β5) .180(.050)* .153(.051)* .138(.052)* .080(.062)
X6 Pgens (β6) .084(.049) .075(.048) .072(.050) .078(.059)
Teacher Level Demographics
X7 Teacht (β7) -.043(.048) -.035(.048) -.033(.047)
X8 Texpt (β8 ) .001(.082) -.006(.081) .007(.079)
X9 Tcurt (β9) -.051(.058) -.043(.058) -.028(.057)
X10 Taget (β10) .029(.075) .030(.075) .023(.073)
X11 Tgent (β11) .066(.050) .064(.049) .065(.048)
X12TSCPt (β12) .063(.050) .086(.051) .010(.053)
School Level Leadership
X13 VIs (β13) .065(.081) .040(.081) .022(.088)
X14 ICs (β14 ) .053(.079) .093(.086) .030(.093)
X15 LFs (β15) -.056(.070) -.008(.067) .020(.096)
School Level School Learning Environment
X16 SSs (β16) -.082(.062)
X17 AFFs (β17) .097(.059) .103(.045)*
X18 PIs (β18) na na
X19 CENs (β19) -.051(.079)
X20 INNs (β20) .229(.079)* .246(.045)*
X21 RAs (β21) .101(.062)
X22 AOs (β22) -.053(.066)
Random:
u0j (School-level) .048(.030) .025(.026) .011(.022) .005(.021) .000(.000) .000(.000) .000(.000)
e0ij (Teacher-level) .921(.054)* .920(.064)* .914(.064)* .911(.063)* .901(.060)* .860(.057)* .888(.059)*
% of Variance Explained 2.5% 4.5% 5.5% 7.0% 11.2% 8.4%
-2 (Log-Likelihood) 1281.432 1272.645 1263.908 1259.734 1252.002 1230.884 1245.185
Note. IGLS solutions shown; fitted parameter estimates with standard errors in parentheses. * Statistically significant beyond the p < .05 level by univariate two tailed test: i.e., the
parameter estimate is greater than twice its standard error; “na” indicates that the associated
school learning environment variable was omitted from the analysis.
138
these two blocks of explanatory variables had a statistically significant influence on
professional interest scores.
Figure 5.10. Residuals of school level professional interest (PIs) scores, ranked for
52 schools, showing 95% confidence intervals.
The addition of school-average level school learning environment
explanatory variables is indicated in Multilevel Model 6 (Table 5.3). Initially, only
one factor, innovation (INNs) was indicated as having a positive, statistically
significant influence on teachers’ professional interest scores. This suggests that
innovation (INNs) is a good catalyst in developing professional interest (PIs) among
teachers. Together, the four blocks of explanatory variables accounted for 11.2% of
the residual variance in teachers’ professional interest scores, with all of the recorded
variance indicated at the teacher level. Reduced Multilevel Model 6 indicated the
effect of removing all non-significant explanatory variables from the model. Two
explanatory variables are indicated as accounting for 8.4% of the recorded variance
in teachers’ professional interest scores, innovation (INNs) and affiliation (AFFs).
This indicates that a sense of collegiality and the press of new methods or the
introduction of new technology are conducive to creating a sense of professional
interest among teachers. Both of these school level explanatory variables are
indicated as having a positive, statistically significant influence on teacher
139
perceptions of professional interest. None of the three leadership explanatory
variables, vision (VIs), individualised consideration (ICs) and laissez-faire leadership
(LFs) were indicated as having a statistically significant influence on professional
interest scores.
There are two factors that influenced teachers’ perceptions of professional
interest (PIs) scores. It was unexpected that none of the three school leadership
variables demonstrated any statistically significant influence on teacher professional
interest (PIs) scores (Hypotheses 1.1 and 1.2). Rather, two school learning
environment factors, affiliation (AFFs) and innovation (INNs) were indicated as
influencing teacher perception of professional interest (PIs) scores.
Centralisation. Centralisation refers to the degree to which teachers perceive
that decisions are made by individuals or by small groups of people, and whether or
not they can influence the decision-making processes in their schools. Table 5.4
indicates the results of fitting multilevel regression models to teachers’ centralisation
scores.
Multilevel Model 1 (Table 5.4) indicates the result of fitting a two level
variance components model to teachers’ centralisation scores. Only variance
recorded at the teacher level was indicated as being statistically significant. Figure
5.11 shows the ranked residuals plotted for each of the school level centralisation
data sets. Three schools were indicated as departing significantly from the overall
average line as predicted by the fixed parameter, and are indicated as lying either
above or below the average line.
Multilevel Model 2 (Table 5.4) indicates the amount of variance that can be
attributed to the three leadership explanatory variables. Initially, only individualised
consideration (ICs) is shown to be statistically significant. The three leadership
factors together accounted for 5.7% of the variation recorded in teachers’
centralisation scores.
140
Table 5.4
Variation in Teachers’ Centralisation Scores (N = 458) in 52 Schools
Explanatory Variables
Multilevel Model 1
MultilevelModel 2
MultilevelModel 3
MultilevelModel 4
Multilevel Model 5
Multilevel Model 6
ReducedModel 6
Fixed:
Constant
(β0jX0): .001(.058) .000(.023) -.001(.052) -.002(.048) -.001(.042) .000(.040) .000(.041)
School Level Demographics
X1 Teachs (β1) -.050(.055) .016(.052) .034(.047) -.014(.051)
X2 Texps (β2 ) .078(.091) .069(.091) .062(.084) .055(.102)
X3 Ssizs (β3) -.082(.054) -.075(.050) -.043(.050) -.033(.050)
X4 Tages (β4) -.128(.087) -.097(.086) -.078(.079) -.008(.087)
X5 Tgens (β5) .103(.055) .085(.053) .075(.049) -.004(.060)
X6 Pgens (β6) -.056(.054) -.075(.050) -.064(.048) -.083(.055) -.135(.042)*
Teacher Level Demographics
X7 Teacht (β7) -.369(.045)* -.359(.045)* -.353(.044)* -.353(.041)*
X8 Texpt (β8 ) .061(.076) .053(.075) .061(.074)
X9 Tcurt (β9) -.124(.055)* -.119(.054)* -.098(.053) -.100(.042)*
X10 Taget (β10) -.061(.070) -.059(.069) -.065(.068)
X11 Tgent (β11) .018(.046) .016(.046) .016(.045)
X12TSCPt (β12) .006(.049) .038(.048) -.024(.050)
School Level Leadership
X13 VIs (β13) .132(.076) .092(.077) .071(.079)
X14 ICs (β14 ) .183(.076)* .138(.082) .082(.085)
X15 LFs (β15) .094(.066) .067(.064) .104(.086)
School Level School Learning Environment
X16 SSs (β16) .008(.060)
X17 AFFs (β17) .022(.058)
X18 PIs (β18) -.009(.070)
X19 CENs (β19) na na
X20 INNs (β20) .238(.066)* .256(.042)*
X21 RAs (β21) .069(.058)
X22 AOs (β22) -.063(.060)
Random:
u0j (School-level) .062(.033) .011(.023) .032(.027) .028(.023) .004(.018) .000(.000) .000(.000)
e0ij (Teacher-level) .925(.065)* .920(.064)* .925(.065)* .780(.054)* .777(.054)* .745.049)* .765(.051)
% of Variance
Explained 5.7% 3.0% 18.1% 20.9% 24.5% 22.5%
-2 (Log-Likelihood) 1287.616 1266.669 1277.911 1200.028 1186.602 1165.011 1177.050
Note. IGLS solutions shown; fitted parameter estimates with standard errors in parentheses. * Statistically significant beyond the p < .05 level by univariate two tailed test: i.e., the
parameter estimate is greater than twice its standard error; “na” indicates that the associated
school learning environment variable was omitted from the analysis.
141
Figure 5.11. Residuals of school level centralisation (CENs) scores, ranked for 52
schools, showing 95% confidence intervals.
Multilevel Models 3 to 6 (Table 5.4) show the progressive result of fitting
four blocks of explanatory variables to teachers’ centralisation scores. No school-
average level demographic variable had any statistically significant effect on
teachers’ centralisation scores. The addition of teacher level demographic variables
(Multilevel Model 4) indicates that teacher position (Teacht; whether administrative
or teaching staff) and teacher’s current length of time in the school (Tcurt) both had a
statistically significant negative influence on teachers’ centralisation scores. Teacher
position (Teacht) had a statistically significant, large negative influence on
centralisation. Non-executive staff indicated their belief that they have little effect in
the decision-making processes within their schools. These two blocks of school
average level and teacher level demographic explanatory variables accounted for
18.1% of the recorded variance, with the majority of this occurring at the teacher
level.
The addition of the leadership block of three explanatory variables is shown
as Multilevel Model 5 (Table 5.4). None of the three leadership variables were
indicated as having a statistically significant influence on teachers’ perceptions of
142
centralisation. Multilevel Model 6 (Table 5.4) indicates the accumulative result of
adding in the influence of school average school learning environment explanatory
variables. Only one variable, innovation (INNs), was indicated as having a
statistically significant, positive influence on teachers’ centralisation scores.
However, along with the other blocks of variables it accounted for almost one quarter
(24.6%) of the recorded variance, all of this recorded at the teacher level.
Reduced Multilevel Model 6 (Table 5.4) indicates the result of removing all
non-significant explanatory variables from the analysis. Four explanatory variables
are indicated as having a negative, statistically significant effect on teachers’
centralisation scores. Two of these are at the teacher level: teacher position (Teacht)
and teacher’s length of time in their current school (Tcurt). Principal’s gender
(Pgens) at the school-average level was also indicated as having a statistically
significant, negative influence on teachers’ perceptions of centralisation. This
indicated that female principals were perceived as influencing perceptions of
centralisation to a lesser extent than male principals. Only one explanatory variable,
innovation (INNs) was recorded as having a positive, statistically significant effect
on teachers’ centralisation scores. Innovation (INNs) was found to enhance teachers’
perception of centralisation. This is suggestive that teachers perceive that schools
with high levels of innovation (INNs) also demonstrate high levels of centralisation
(CENs). Teachers perceive that decision-making occurs at the principal and senior
executive level, and that individual teachers, therefore, don’t have much say in what
happens. Together, these four explanatory variables account for 22.5% of the
variance, all of the residual variance being at the teacher level. Interestingly,
individualised consideration (ICs) as a leadership explanatory variable was initially
shown to be statistically significant as having an influence on teachers’ perceptions
of centralisation (CENs). The addition of the other blocks of explanatory variables
negated the statistical significance of individualised consideration (ICs). This
suggests that the effect of individualised consideration (ICs) is mediated by the
inclusion of other explanatory variables. Again, this is a product of the causal
ordering of explanatory variables, which can have the effect of suppressing the
influence of variables, depending on the order in which they are introduced to the
analysis.
143
It was unexpected that none of the three leadership variables considered in
this study demonstrated any statistically significant influence on teacher perception
of centralisation (CENs) (Hypothesis 1.2). However, it was also demonstrated that
principal’s gender (Pgens) influenced teacher perceptions of centralisation (CENs),
along with two other teacher level variables related to position (Teacht) and length of
tenure (Tcurt) in their schools (Research Question 3)
Innovation. Table 5.5 shows the results of fitting multilevel regression
models to teachers’ innovation scores. Innovation is defined as the degree to which
teachers perceive the school is in favour of planned change and experimentation.
Multilevel Model 1 (Table 5.5) indicates the result of fitting a two level
variance components model to teachers’ innovation scores. Variance at both the
school level and teacher level was indicated as being statistically significant. Figure
5.12 shows the ranked residuals plotted for each of the school level innovation data
sets. Several schools were shown as departing significantly from the overall average
line as predicted by the fixed parameter, and were indicated as lying either above or
below the average line.
Multilevel Model 2 (Table 5.5) indicated the amount of variance that can be
attributed to the three leadership explanatory variables. Collectively, vision (VIs),
individualised consideration (ICs) and laissez-faire leadership (LFs) accounted for
over 5% of the recorded variance, and this was statistically significant. However,
individually, none of the leadership explanatory variables were demonstrated to have
any statistical significance. Multilevel Models 3 to 6 (Table 5.5) indicate the results
of fitting four blocks of explanatory variables to account for the variation in teachers’
innovation scores. Multilevel Model 3 (Table 5.5) shows the combined effect of
adding a set of school-average level demographic variables to the analysis. The
demographic explanatory variable of teacher gender (Tgens) was indicated as having
a statistically significant, negative influence on teacher perceptions of innovation
(INNs) scores. In Multilevel Model 4 (Table 5.5), the combined effects of a set of
teacher level variables were added to the analysis. The demographic variable of
teacher position (Teacht) was initially indicated as having a negative, statistically
significant effect on teachers’ innovation (INNs) scores.
144
Table 5.5
Variation in Teachers’ Innovation Scores (N = 458) in 52 Schools Explanatory Variables
Multilevel Model 1
Multilevel Model 2
MultilevelModel 3
MultilevelModel 4
Multilevel Model 5
Multilevel Model 6
ReducedModel 6
Fixed:
Constant
(β0jX0): -.014(.072) -.006(.064) -.010(.063) -.012(.062) -.008(.057) -.003(.040) -.005(.043)
School Level Demographics
X1 Teachs (β1) .004(.065) .040(.064) .052(.061) .124(.050)*
X2 Texps (β2 ) .087(.103) .099(.107) .088(.102) -.175(.100) -.085(.043)*
X3 Ssizs (β3) -.041(.063) -.038(.062) -.001(.065) .026(.050)
X4 Tages (β4) -.185(.099) -.206(.102)* -.180(.097) -.006(.088)
X5 Tgens (β5) .182(.065)* .170(.065)* .158(.063)* -.017(.060)
X6 Pgens (β6) .120(.065) .110(.064) -.104(.064) .054(.056) .113(.045)*
Teacher Level Demographics
X7 Teacht (β7) -.191(.045)* -.187(.045)* -.177(.044)* -.170(.042)*
X8 Texpt (β8 ) -.011(.075) -.018(.075) -.016(.074)
X9 Tcurt (β9) -.061(.055) -.059(.055) -.034(.053)
X10 Taget (β10) .059(.069) .061(.069) .056(.068)
X11 Tgent (β11) .013(.046) .012(.046) .011(.045)
X12TSCPt (β12) .000(.054) .028(.053) -.011(.049)
School Level Leadership
X13 VIs (β13) .099(.100) .010(.100) -.092(.082)
X14 ICs (β14 ) .045(.099) .135(.108) -.072(.087)
X15 LFs (β15) -.105(.088) -.023(.085) -.176(.087)*
School Level School Learning Environment
X16 SSs (β16) .071(.060)
X17 AFFs (β17) .002(.058)
X18 PIs (β18) .230(.062)* .230(.050)*
X19 CENs (β19) .258(.058)* .211(.052)*
X20 INNs (β20) na na
X21 RAs (β21) -.063(.058)
X22 AOs (β22) .138(.058)*
Random:
u0j (School-level) .161(.051)* .112(.041)* .103(.039)* .100(.038)* .072(.032) .000(.000) .006(.018)
e0ij (Teacher-level) .811(.057)* .810(.057)* .808(.056)* .766(.054)* .769(.054)* .750(.050)* .769(.054)*
% of Variance
Explained 5.1% 6.3% 10.9% 13.5% 22.8% 20.2%
-2 (Log-Likelihood) 1254.508 1243.417 1239.820 1216.338 1209.840 1167.747 1182.878
Note. IGLS solutions shown; fitted parameter estimates with standard errors in parentheses. * Statistically significant beyond the p < .05 level by univariate two tailed test: i.e., the
parameter estimate is greater than twice its standard error; “na” indicates that the associated
school learning environment variable was omitted from the analysis.
145
Figure 5.12. Residuals of school level innovation (INNs) scores, ranked for 52
schools, showing 95% confidence intervals.
In Multilevel Model 5 (Table 5.5), a block of three leadership explanatory
variables was added to the blocks of school and teacher level demographic variables
to account for the variance in teachers’ innovation scores. None of the vision (VIs),
individualised consideration (ICs) or laissez-faire leadership (LFs) explanatory
variables were demonstrated as having a statistically significant effect on teachers’
innovation (INNs) scores. Multilevel Model 6 (Table 5.5) indicates the addition of
school-average level school learning environment explanatory variables. Three
variables are indicated as having a statistically significant, positive influence on
teachers’ innovation scores, namely: professional interest (PIs), centralisation
(CENs) and achievement orientation (AOs).
Reduced Model 6 (Table 5.5) indicates the result of removing all non-
significant explanatory variables in accounting for variations in teachers’ innovation
scores. Five explanatory variables were indicated as accounting for 20.2% of the
variance in teachers’ innovation scores, the majority of which were statistically
significant at the teacher level. Two variables were indicated to have a statistically
significant, negative influence on teachers’ innovation scores, the school-average
level demographic variable teacher experience (Texps) and teacher level teacher
146
position (Teacht). These negative explanatory variables suggest that younger, more
inexperienced teachers perceive a lesser degree of innovation within their schools
than do older, more experienced teachers. Three explanatory variables were
indicated as having a statistically significant, positive influence on innovation scores,
school-average level principal gender (Pgens), school-average level professional
interest (PIs) and centralisation (CENs). These explanatory variables suggest that
teachers perceive a positive relation in their involvement with decision-making and
professionalism and the implementation of innovation within their schools.
It was unexpected that none of the three leadership variables under
consideration in this phase of the study demonstrated any statistically significant
influence on teachers’ innovation (INNs) scores (Hypothesis 1). However, several
antecedent variables indicated statistically significant effects on teachers’ perception
of innovation (INNs) scores (Research Question 3).
Resource Adequacy. Resource adequacy is defined as teachers’ perceptions
of the level of support personnel, facilities and resources that are available to them.
Table 5.6 shows the results of fitting multilevel regression models to teachers’
resource adequacy scores. Multilevel Model 1 (Table 5.6) shows the result of fitting
a two level variance component model to teachers’ resource adequacy scores.
Although the random part of the model indicates that variance has occurred at both
school level and teacher level, most of the statistically significant variance is at the
teacher level. Figure 5.13 shows the ranked residuals plotted for each of the school
level resource adequacy scores. Only two schools were indicated as departing
significantly from the overall average line.
147
Table 5.6
Variation in Teachers’ Resource Adequacy Scores (N = 458) in 52 Schools
Explanatory Variables
Multilevel Model 1
MultilevelModel 2
MultilevelModel 3
MultilevelModel 4
Multilevel Model 5
Multilevel Model 6
Reduced Model 6
Fixed:
Constant
(β0jX0): -.012(.055) -.010(.052) -.010(.052) -.011(.053) -.010(.051) -.010(.042) -.011(.045)
School Level Demographics
X1 Teachs (β1) .095(.055) .123(.057)* .107(.056)* .104(.053)*
X2 Texps (β2 ) .031(.090) -.028(.098) -.022(.096) -.286(.102)*
X3 Ssizs (β3) .049(.053) .053(.055) .048(.059) .041(.052)
X4 Tages (β4) .018(.086) .066(.093) .063(.091) .178(.089)*
X5 Tgens (β5) .047(.054) .075(.058) .075(.058) -.082(.059)
X6 Pgens (β6) -.072(.054) -.080(.056) -.107(.058) -.120(.054)*
Teacher Level Demographics
X7 Teacht (β7) -.210(.046)* -.212(.046)* -.218(.046)* -.218(.043)*
X8 Texpt (β8 ) .047(.077) .050(.077) .056(.076)
X9 Tcurt (β9) .018(.056) .021(.056) .030(.054)
X10 Taget (β10) -.073(.071) -.075(.071) -.079(.070)
X11 Tgent (β11) -.064(.047) -.064(.047) -.062(.046)
X12TSCPt (β12) .074(.052) .057(.052) .015(.052)
School Level Leadership
X13 VIs (β13) -.096(.083) -.111(.091) -.185(.080)* -..232(.065)*
X14 ICs (β14 ) -.055(.081) -.053(.098) -.161(.087)
X15 LFs (β15) -.126(.072) -.153(.077)* -.330(.078)* -.201(.065)*
School Level School Learning Environment
X16 SSs (β16) .225(.054)* .165(.046)*
X17 AFFs (β17) .079(.058) .127(.049)*
X18 PIs (β18) .152(.071)*
X19 CENs (β19) .122(.075)
X20 INNs (β20) -.155(.085)
X21 RAs (β21) na na
X22 AOs (β22) .132(.061)*
Random:
u0j (School-
level) .048(.030) .036(.027) .035(.027) .049(.028) .038(.026) .000(.000) .010(.020)
e0ij (Teacher-
level) .886(.062)* .887(.062)* .883(.062)* .817(.057)* .815(.057)* .798(.053)* .832(.058)*
% of Variance
Explained 1.2% 1.7% 7.2% 8.7% 14.6% 9.9%
-2 (Log-
Likelihood) 1264.190 1260.180 1257.699 1228.760 1223.855 1196.488 1221.128
Note. IGLS solutions shown; fitted parameter estimates with standard errors in parentheses. * Statistically significant beyond the p < .05 level by univariate two tailed test: i.e., the
parameter estimate is greater than twice its standard error; “na” indicates that the associated
school learning environment variable was omitted from the analysis. .
148
Figure 5.13. Residuals of school level resource adequacy (RAs) scores, ranked for 52
schools, showing 95% confidence intervals.
Multilevel Model 2 (Table 5.6) indicates the result of fitting regression
models using only the three leadership factors as explanatory variables. At this stage
in the analysis, none of the three leadership variables are demonstrated to have a
statistically significant influence on teachers’ resource adequacy scores. The random
part of the model indicates that while teacher level effects are statistically significant,
only a small part of the total variance (1.2%) can be attributed to those leadership
factors. Multilevel Model 3 (Table 5.6) shows the result of fitting regression models
using a block of six school-average level demographic variables. None of these six
explanatory variables was indicated to be statistically significant. Multilevel Model 4
(Table 5.6) shows the accumulative results of fitting a block of six teacher level
demographic explanatory variables to account for the variance in teachers’ resource
adequacy scores. Only teacher position (Teacht) was indicated as having a negative
but statistically significant influence on teachers’ resource adequacy scores. The
addition of the teacher level explanatory variables caused the corresponding school
level explanatory variable of teacher position (Teachs) to become statistically
significant but positive in its influence of teachers’ resource adequacy scores.
149
Together, these two blocks of explanatory variables in Multilevel Model 4 (Table
5.6) accounted for 7.2% of the recorded variance. While variance was indicated at
both school and teacher levels, only variance residual at the teacher level was
demonstrated to be statistically significant. The school level variance shown in the
random part of the model was explained by the explanatory variables, hence the
residual variance was non-significant.
Multilevel Model 5 (Table 5.6) adds in the effects of the school level
leadership explanatory variables vision (VIs), individualised consideration (ICs) and
laissez-faire leadership (LFs). At this point in the analysis, laissez-faire leadership
(LFs) is indicated as having a negative, statistically significant effect on teachers’
resource adequacy scores along with teacher level teacher position on staff (Teacht).
Schools in which the staff perceive their principal’s behaviour as being highly
laissez-faire also perceive a reduction in resource availability. School-average level
teacher position on staff (Teachs) is also indicated as having a statistically
significant, positive influence on teachers’ resource adequacy scores. Together, the
three blocks of explanatory variables considered so far in the analysis accounted for
8.7% of the variance recorded, the large majority of which was statistically
significant at the teacher level. Multilevel Model 6 (Table 5.6) examines the result of
adding into the analysis the fourth block of school level school learning environment
explanatory variables. At this point, five explanatory variables are indicated as
having a positive, statistically significant, influence on teachers’ resource adequacy
scores; student supportiveness (SSs), centralisation (CENs), achievement orientation
(AOs), teacher position on staff (Teachs) and teacher ages (Tages) at the school-
average level. Interestingly, the addition of the school learning environment
explanatory variables has caused the leadership variables vision (VIs) and laissez-
faire leadership (LFs) to have a negative, statistically significant influence of
teachers’ resource adequacy scores. Teacher experience (Texps) and principal gender
(Pgens) at the school level, and teacher position on staff at the teacher level (Teacht)
are also indicated as having a statistically significant, negative influence on teachers’
resource adequacy scores. Together, the variables in Multilevel Model 6 explain
14.6% of the recorded variance. Reduced Model 6 removes all non-statistically
significant variables from the analysis, indicates two explanatory variables as having
a positive influence on teachers’ resource adequacy scores, and three explanatory
150
variables as having a negative influence on these scores. Student supportiveness
(SSs) and affiliation (AFFs) at the school level are both indicated as having a
significant positive influence on teachers’ resource adequacy scores. Teacher level
teacher position (Teacht), and school level vision (VIs) and laissez-faire leadership
(LFs) are indicated as having a negative influence on these same scores. These
results suggest that perceptions of resource adequacy are influenced by the seniority
of staff members. Vision (VIs), as a leadership behaviour was also perceived as
having a negative effect as far as the level of resource availability is concerned, and
is perhaps reflective of the fact that many teachers are called on to do more by way
of implementing innovations within their schools with less allocation of resources.
Together, these explanatory variables account for 9.9% of the recorded variance, the
majority of which is statistically significant at the teacher level.
The initial non-significance of the three leadership variables (Model 2) and
subsequent negative significance of vision (VIs) and laissez-faire leadership (LFs)
(Model 6; Reduced Model 6) is acknowledged. The effects of explanatory variables
on the response variable (as shown in Table 5.6) are analogous to a path model, with
the direct effects shown in Models 5 and 6, and the indirect effects mediated through
the demographic variables (Models 3 and 4). Even though the total effect of
leadership (as shown in Model 2) is shown as non-significant, this may well be due
to a suppression effect resulting from the causal ordering of explanatory variables. In
this case, the influence of the direct effects of the leadership variables may have been
offset by the mediating influence of other variables. An explanation as to the non-
significance of the leadership explanatory variables may well lie in the fact that
funding for special projects (and hence perceptions of resource adequacy among
teachers) comes from a centrally administrated authority (regional office). Principals
may well be seen as being under the control of this authority as far as spending on
resources is concerned.
Two of the three leadership explanatory variables, visionary leadership (VIs)
and laissez-faire leadership (LFs) indicated a statistically significant negative
influence on teachers’ perception of resource adequacy (RAs) scores (Hypothesis
1.1). While, as far as laissez-faire leadership (LFs) is concerned, this result is
consistent with predictions, the result for visionary leadership (VIs) is unexpected
151
and counter to predictions made in the transformational leadership literature. One
teacher level antecedent variable, teacher position in their school (Teacht)
demonstrated a statistically significant negative influence on these teachers’ scores
(Research Question 3).
Achievement Orientation. Achievement orientation refers to the degree to
which teachers expect and value high student achievement. Table 5.7 indicates the
results of fitting multilevel regression models to teachers’ achievement orientation
scores. Figure 5.14 shows the ranked residuals plotted for each of the school level
resource adequacy scores. Six schools were indicated as departing significantly from
the overall average line.
Figure 5.14. Residuals of school level achievement orientation (AOs) scores, ranked
for 52 schools, showing 95% confidence intervals.
Multilevel Model 1 (Table 5.7) shows the result of fitting a two level variance
component model to teachers’ achievement orientation scores. The random part of
the model indicates that most of the statistically significant variance is at the teacher
level.
152
Table 5.7
Variation in Teachers’ Achievement Orientation Scores (N = 458) in 52 Schools
Explanatory
Variables MultilevelModel 1
MultilevelModel 2
MultilevelModel 3
MultilevelModel 4
Multilevel Model 5
MultilevelModel 6
Reduced Model 6
Fixed:
Constant
(β0jX0): -.009(.057) -.011(.052) -.011(.045) -.011(.044) -.011(.043) -.011(.042) -.011(.043)
School Level Demographics
X1 Teachs (β1) -.118(.048)* -.130(.048)* -.121(.048)* -.138(.051)* -.148(.044)*
X2 Texps (β2 ) .160(.080)* .065(.088) .062(.086) .166(.101) .184(.047)*
X3 Ssizs (β3) .071(.047) .074(.047) .068(.050) .053(.052)
X4 Tages (β4) .066(.077) .085(.082) .079(.080) .020(.091)
X5 Tgens (β5) .040(.047) .006(.050) .012(.050) .047(.061)
X6 Pgens (β6) -.040(.047) -.033(.047) -.001(.048) .026(.059)
Teacher Level Demographics
X7 Teacht (β7) -.023(.046) -.023(.046) -.019(.046)
X8 Texpt (β8 ) .077(.078) .079(.078) .084(.077)
X9 Tcurt (β9) .098(.056) .089(.055) .095(.055) .148(.046)*
X10 Taget (β10) .007(.072) .009(.072) .007(.071)
X11 Tgent (β11) .138(.048)* .138(.047)* .138(.047)* .120(.044)*
X12TSCPt (β12) .084(.048) .091(.049) .064(.052)
School Level Leadership
X13 VIs (β13) .080(.083) .082(.078) .131(.084)
X14 ICs (β14 ) -.045(.081) .004(.083) .052(.091)
X15 LFs (β15) .168(.072)* .170(.064)* .278(.083)* .146(.047)*
School Level School Learning Environment
X16 SSs (β16) -.127(.059)*
X17 AFFs (β17) .022(.060)
X18 PIs (β18) -.061(.073)
X19 CENs (β19) -.107(.075)
X20 INNs (β20) .174(.081)* .104(.048)*
X21 RAs (β21) .111(.059)
X22 AOs (β22) na na
Random:
u0j (School-level) .062(.057) .035(.027) .003(.020) .006(.020) .000(.000) .000(.000) .000(.000)
e0ij (Teacher-level) .883(.062)* .886(.062)* .880(.061)* .837(.058)* .827(.055)* .809(.053)* .840(.055)
% of Variance
Explained 2.5% 6.6% 10.8% 12.5% 14.4% 11.1%
-2 (Log-Likelihood) 1267.101 1259.277 1243.024 1221.234 1212.933 1202.785 1219.749
Note. IGLS solutions shown; fitted parameter estimates with standard errors in parentheses. * Statistically significant beyond the p < .05 level by univariate two tailed test: i.e., the
parameter estimate is greater than twice its standard error; “na” indicates that the associated
school learning environment variable was omitted from the analysis. .
153
Multilevel Model 2 (Table 5.7) examines the influence of the three leadership
variables alone in accounting for the variance in teachers’ achievement orientation
scores. While together these three explanatory variables account for 2.5% of the
recorded variance in teachers’ scores, only laissez-faire leadership (LFs) is indicated
as having a positive yet statistically significant influence on teachers’ achievement
orientation scores.
Multilevel Models 3 to 6 (Table 5.7) show the cumulative addition of four
blocks of explanatory variables to account for the variation in teachers’ achievement
orientation scores. Multilevel Model 3 (Table 5.7) indicates the effect of six school-
average level demographic explanatory variables. The random part of the model
indicates that this block of explanatory variables is responsible for 6.6% of the
recorded variation, the majority of which has occurred at the school level. The large
change in the log likelihood statistic between Multilevel Model 1 and Multilevel
Model 3 also indicates its significance. Two factors in this block are indicated as
having statistical significance, school-average level teacher position (Teachs) having
a negative influence and school-average level teacher experience (Texps) a positive
influence on teachers’ achievement orientation scores. Schools where the school
average level of experience is higher (Texps) may be more likely to focus on
conveying to students the importance of academic achievement, whereas relatively
inexperienced staff may focus on mastery of content.
Multilevel Model 4 adds six teacher level demographic explanatory variables
into this analysis. These two blocks (teacher level and school level demographic
variables) explain 10.8% of the recorded variance, and a large statistically significant
portion of this is at the teacher level. Within the teacher level demographic block of
explanatory variables, teacher gender (Tgent) is indicated as being statistically
significant and having a positive influence on teachers’ perception of achievement
orientation scores. This result indicates that teachers’ perceptions are that female
teachers are more orientated towards encouraging achievement in their students than
are male teachers.
Multilevel Model 5 (Table 5.7) includes in this analysis the accumulative
effect of three leadership explanatory variables, vision (VIs), individualised
154
consideration (ICs) and laissez-faire leadership (LFs), along with the influence of the
previous two blocks of explanatory variables. These three blocks account for 12.5%
of the recorded variance, with reductions in variance components occurring at both
the school and the teacher level. Laissez-faire leadership (LFs) is indicated as having
a positive statistically significant influence on teachers’ achievement orientation
scores. Multilevel Model 6 (Table 5.7) adds in the effect of six school level school
learning environment explanatory variables into this analysis. Initially, two variables,
student supportiveness (SSs) and innovation (INNs) are indicated as having a
statistically significant negative and positive effect respectively on teachers’
achievement scores. The four blocks of explanatory variables account for 14.4% of
the recorded variance.
Reduced Model 6 removes the influence of all non-significant explanatory
variables from the analysis, the remaining explanatory variables accounting for
11.1% of the variance in teachers’ achievement orientation scores. All of the residual
variation in these scores is at the teacher level. The results indicated a negative,
statistically significant relation between teacher position (Teachs) at the school level
and teachers’ perception of achievement orientation. However, five explanatory
variables were indicated as having a statistically significant enhancing effect on
teachers’ perceptions of the push for achievement orientation. School level teacher
experience (Texps), teacher level teacher time in current school (Tcurt) and teacher
gender (Tgent) were all indicated as enhancing teachers’ perceptions of achievement
orientation. Innovation (INNs) was also indicated by teachers as a catalyst for
promoting an achievement orientation; this finding is suggestive that change
implementation and innovation fostered a greater push for achievement. Laissez-faire
leadership (LFs) had a statistically significant and positive impact on perceptions of
achievement orientation, indicating that schools where principals took a “hands off”
approach to leadership were more likely to have a higher achievement orientation. A
possible explanation for this result is the loose coupling that occurs between teachers
and principals (Weick, 1976). Teachers perceive that principals who take a hands-off
approach in their leadership style are less likely to interfere in their classrooms,
allowing them to be more autonomous.
155
Laissez-faire leadership (LFs) was the only leadership explanatory variable
that demonstrated any statistically significant effect on teachers’ perception of
achievement orientation (AOs) sores. This finding contradicts both Hypothesises 1.1
and 1.2. Four antecedent variables, two at the teacher level and two at the school
level were also indicated as having a statistically significant effect on these teachers’
scores (Research Question 3).
Hypothesis 1.1: Results
It has been hypothesised, largely on the basis of previously conducted
research (Bass, 1985a; Bass & Avolio, 1997), that teachers who have experienced a
transformational leadership style will display higher scores compared to teachers
who have experienced transactional or laissez-faire leadership styles for the
effectiveness of the school learning environment in student supportiveness, teacher
affiliation, professional interest, achievement orientation, centralisation (where
decision-making occurs), innovation and resource adequacy.
A significant positive result is indicative that teachers scored higher on a
particular leadership explanatory variable. Alternatively, a significant result in the
negative direction suggests that teachers scored lower on a particular leadership
explanatory variable. A non-significant result implies that teachers’ perceptions of
leadership behaviour as an explanatory factor accounting for variation in school
learning environment scores was not significantly different to other explanatory
factors.
The leadership style behaviour vision (VIs) was demonstrated as having a
statistically significant effect on the school learning environment variable of resource
adequacy (RAs). An analysis of the data gathered across 458 staff in 52 schools as
part of this study, indicated that the visionary leadership (VIs) behaviour of
principals had a statistically significant, negative influence on only one of the seven
school learning environment constructs, resource adequacy (RAs). Teachers perceive
that an increase in visionary leadership (VIs) behaviour is associated with a decrease
in resource adequacy (RAs). It is likely that teachers perceive an increase in
visionary behaviour of principals will result in fewer resources for them to complete
156
their teaching duties. As public schools tend to be centrally funded, a principal’s
visionary project may well translate into fewer resources being available for other
areas and programs in the school. Again, the lament of some teachers interviewed in
the qualitative phase of this study was that central educational authorities expected
the achievement of more outcomes with the provision of fewer resources.
In three of the school learning environment variables, professional interest
(PIs), centralisation (CENs) and innovation (INNs), the non-significant result
suggests that teachers could not differentiate between the effects of the visionary
(VIs), individualised consideration (ICs) and the laissez-faire (LFs) leadership. For
three of the remaining school learning environment variables, student supportiveness
(SSs), affiliation (AFFs) and achievement orientation (AOs), the leadership
behaviours of individualised consideration (ICs) and laissez-faire leadership (LFs)
were indicated as having a more positive influence than that of vision (VIs).
The evidence presented indicated that teachers who have experienced a
transformational leadership style have displayed either no difference or significantly
lower scores compared to teachers who have experienced transactional or laissez-
faire leadership styles for the effectiveness of the school-learning environment.
Therefore, on the basis of the multilevel analysis undertaken, Hypothesis 1.1 is
rejected.
Hypothesis 1.2: Results
Hypothesis 1.2 postulates that teachers who have experienced a transactional
leadership style will display higher scores when compared to teachers who have
experienced a laissez-faire leadership style for the effectiveness of the school
learning environment.
The school learning environment variable of affiliation (AFFs) was shown as
being influenced by the individualised consideration (ICs) leadership behaviour of
principals. However, affiliation (AFFs) also demonstrated the influence of laissez-
faire leadership (LFs) style behaviour to a greater extent than it does the influence of
the individualised consideration (ICs) leadership behaviour. Again, for three of the
157
remaining school learning environment variables: professional interest (PIs),
centralisation (CENs) and innovation (INNs), the non significant result suggests that
there was no difference between the effects of the transformational vision (VIs),
transformational/transactional hybrid individualised consideration (ICs) and the
laissez-faire (LFs) explanatory variable. In the remaining three school learning
environment variables (student supportiveness [SSs], resource adequacy [RAs] and
achievement orientation [AOs]), both visionary leadership (VIs) and laissez-faire
leadership (LFs) behaviour was indicated as having a more positive influence than
that of individualised consideration (ICs). As far as visionary leadership (VIs)
behaviour is concerned, this result is consistent with predictions made in the
transformational leadership literature. However, in the case of laissez-faire leadership
(LFs), this result is not consistent with predictions made in this same literature.
The evidence presented in the multilevel analysis of the data gathered
indicates that teachers who have experienced an individualised consideration (ICs)
leadership style have displayed lower scores compared to teachers who have
experienced visionary or laissez-faire leadership styles for the effectiveness of the
school learning environment. Therefore, although there is some support for the
positive influence of individualised consideration (ICs) leadership style as far as
affiliation (AFFs) is concerned, on the basis of the multilevel analysis undertaken,
Hypothesis 1.2 is rejected.
Hypothesis 1.1. and 1.2. Summary
Based on the analysis presented, Hypothesis 1.1 and Hypothesis 1.2 are
rejected. The effect of principal’s leadership styles on school learning environment
variables is quite unexpected, and contrary to the transformational leadership
literature. The literature, and reported earlier studies (Bass, 1985a; Bass & Avolio,
1997; Kirby, King & Paradise, 1992) suggested that transformational leadership
modified the contextual environment so as to maximise the impact of the leadership.
However, in this study, visionary leadership (VIs) and individualised consideration
(ICs) behaviour were indicated as being less influential than laissez-faire (LFs) style
leadership behaviour.
158
An analysis of the data gathered in the first phase of this study indicated that
it was the laissez-faire (LFs) leadership behaviours that had the most statistically
significant influence on school learning environment variables. The laissez-faire
(LFs) leadership style behaviours were demonstrated to have a statistically
significant positive relation to student supportiveness (SSs), affiliation (AFFs) and
achievement orientation (AOs), and a statistically significant negative relation to
resource adequacy (RAs).
Teachers perceived that an increase in principals’ laissez-faire (LFs)
leadership behaviour would positively impact student supportiveness (SSs),
affiliation (AFFs) and achievement orientation (AOs). It seems possible that as
teachers perceive laissez-faire style behaviour from their principal, their commitment
to the principal will be substituted by commitment to faculty groups. Teachers
perceive that “hands off” or non-existent leadership on the part of the principal has
the effect of encouraging them to find reward and motivation from other sources,
presumably by looking for leadership support from the faculty head and from faculty
groupings. Kerr and Jermier (1978) suggest that in situations where team members
are highly trained and dysfunctional leadership occurs, there tends to be less reliance
on supervisory structures.
Laissez-faire leadership (LFs) behaviour was also indicated as having a
statistically significant negative impact on resource adequacy (RAs). Teachers
perceived that as laissez-faire (LFs) style behaviour of the principal increased, the
availability and adequacy of resources (RAs) in their schools decreased. It is likely
that teachers saw the principal’s lack of decision-making behaviour as adversely
affecting the quantity of resources available for them to complete their teaching
assignments. Alternatively, this relation could be interpreted as teachers’ frustration
with indecisive leadership because of missed opportunities to gain teaching
resources.
159
Results: Hypothesis 2.1 and 2.2: The Relation of Transformational
and Transactional Leadership Behaviour and Teachers’ Satisfaction
with their Leader, Teachers’ Self-Perceived Effectiveness, Influence
and Control
Overview
Hypothesis 2.1 predicted that teachers who have experienced a
transformational leadership style will display higher scores compared to teachers
who have experienced transactional or laissez-faire leadership styles for satisfaction
with their leader, and ratings of their effectiveness, influence and control as teachers.
A series of four multilevel models was employed to test this hypothesis. Hypothesis
2.2 predicted that teachers who have experienced a transactional leadership style will
display higher scores compared to teachers who have experienced a laissez-faire
leadership style for satisfaction with their leader, and ratings of their effectiveness,
influence and control as teachers.
Analysis
As indicated earlier, multilevel modelling is a technique that addresses the
problems inherent with the nested structure of much educational data (Goldstein et
al., 1998; Keeves & Sellin, 1988; Rowe, 1995), including the data gathered in this
study. A series of four multilevel models was used to examine teacher responses,
both at the school and individual levels.
Global Satisfaction with Leadership. Table 5.8 below shows the result of
fitting multilevel regression models to teachers’ global satisfaction with leadership
(SAT) scores by adding four blocks of explanatory variables. Global satisfaction
with leadership can be defined as the perception of overall follower willingness to
give extra effort, follower satisfaction with leadership and follower perceptions of
leadership effectiveness.
160
Table 5.8
Variation in Teachers’ Global Satisfaction with Leadership Scores (N = 458) in 52 Schools Explanatory Variables
Multilevel Model 1
MultilevelModel 2
MultilevelModel 3
Multilevel Model 4
Multilevel Model 5
Multilevel Model 6
Reduced Model 6
Fixed:
Constant
(β0jX0): -.014(.079) -.002(.039) -.018(.075) -.022(.071) -.002(.037) -.002(.037) -.002(.037)
School Level Demographic
X1 Teachs (β1) -.100(.076) -.048(.073) -.028(.042) -.013(.048)
X2 Texps (β2 ) -.001(.116) .086(.115) .019(.075) -.094(.096)
X3 Ssizs (β3) -.098(.074) -.099(.070) -.010(.044) .004(.046)
X4 Tages (β4) -.101(.113) -.114(.110) .015(.070) .093(.080)
X5 Tgens (β5) .086(.076) .053(.073) -.006(.044) -.068(.055)
X6 Pgens (β6) .035(.077) .014(.073) -.026(.042) -.060(.052)
Teacher Level Demographic
X7 Teacht (β7) -.210(.043)* -.197(.040)* -.201(.040)* -.185(.038)*
X8 Texpt (β8 ) -.059(.071) -.079(.068) -.076(.068)
X9 Tcurt (β9) -.179(.053)* -.168(.048)* -.164(.048)* -.197(.038)*
X10 Taget (β10) .018(.065) .022(.063) .020(.062)
X11 Tgent (β11) .022(.043) .020(.041) .020(.041)
X12 TSCPt (β12) -.072(.054) .004(.043) -.017(.046)
School Level Leadership
X13 VIs (β13) .082(.064) .050(.068) .033(.075)
X14 ICs (β14 ) .345(.062)* .310(.072)* .262(.080)* .361(.049)*
X15 LFs (β15) -.150(.055)* -.169(.056)* -.209(.082)* -.159(.049)*
School Level School Learning Environment
X16 SSs (β16) .055(.055)
X17 AFFs (β17) -.007(.053)
X18 PIs (β18) .155(.064)*
X19 CENs (β19) .019(.068)
X20 INNs (β20) -.070(.077)
X21 RAs (β21) -.033(.054)
X22 AOs (β22) .010(.057)
Random:
u0j (School-
level) .227(.064)* .000(.000) .189(.056)* .169(.050)* .000(.000) .000(.000) .000(.000)
e0ij (Teacher-
level) .763(.053)* .711(.047)* .761(.053)* .677(.047)* .634(.042)* .625(.041)* .639(.042)*
% of Variance
Explained 28.2% 4.0% 14.5% 35.9% 36.9% 35.5%
-2 (Log-
Likelihood) 1239.846 1143.639 1232.416 1179.259 1090.984 1084.422 1094.752
Note. IGLS solutions shown; fitted parameter estimates with standard errors in parentheses. * Statistically significant beyond the p < .05 level by univariate two tailed test: i.e., the
parameter estimate is greater than twice its standard error.
161
Multilevel Model 1 (Table 5.8) indicates the results of fitting a two-level
variance components model to teachers’ global satisfaction with leadership scores.
The random part of the model indicates that variance at both the school and teacher
level was found to be statistically significant. Figure 5.15 shows the ranked residuals
plotted for each of the school level global satisfaction with leadership data sets.
Many schools were indicated as departing significantly from the overall average
line. Hence, results shown in both Model 1 (Table 5.8) and Figure 5.15 indicate that
there are substantial school-to-school differences in global satisfaction with
leadership.
Figure 5.15. Residuals of school level teachers’ perception of global satisfaction with
leadership (SAT) scores, ranked for 52 schools, showing 95% confidence intervals.
Multilevel Model 2 (Table 5.8) indicates the amount of variance that can be
attributed to the three leadership explanatory variables, vision (VIs), individualised
consideration (ICs) and laissez-faire leadership (LFs), without considering the
confounding effects of the other explanatory variables. Individualised consideration
(ICs) is indicated as having a large, statistically significant positive influence on
teacher perceptions of global satisfaction with leadership (SAT), whereas laissez-
faire leadership (LFs) is indicated as having a statistically significant negative
influence on these teacher perceptions. Vision (VIs) is not indicated as having a
162
statistically significant influence on teacher perceptions of global satisfaction with
leadership (SAT). The random part of the Multilevel Model 2 (Table 5.8) indicates
these three explanatory variables account for 28.2% of the explained variance, and
that residual variance is indicated at both the school and the teacher levels.
Multilevel Model 3 (Table 5.8) indicates the amount of variance that can be
attributed to the school-average level school demographic explanatory variables.
While collectively these six explanatory variables accounted for 4% of the recorded
variance, none of these explanatory variables were demonstrated to have any
statistical significance. The change in the log likelihood statistic between Multilevel
Models 1 and 3 also indicated the non-significance of these explanatory variables.
Multilevel Model 4 (Table 5.8) indicates the result of fitting a block of six teacher
level explanatory variables to account for the variation in teachers’ global
satisfaction with leadership scores. The demographic explanatory variables of
teacher position (Teacht) and teacher current time in the school (Tcurt) were
indicated as having a statistically significant, negative influence on teachers’
perception of global satisfaction with leadership (SAT) scores (Multilevel Model 4,
Table 5.8). This result suggests that more senior and longer serving teachers are less
satisfied with leadership than are younger teachers. The random part of the model
indicates that the addition of this block of explanatory variables explains 14.5% of
the variance. In addition, the random part of the model indicates the residual variance
(unexplained variance) at the school level is also statistically significant.
In Multilevel Model 5 (Table 5.8), a block of three leadership explanatory
variables has been added to school and teacher level demographic variables to
account for the variance in teachers’ global satisfaction with leadership scores. Both
individualised consideration (ICs) and laissez-faire leadership (LFs) explanatory
variables were demonstrated as being statistically significant. The addition of this
block of explanatory variables, along with the teacher level and school level
demographic variables, accounted for 35.9% of the recorded variance. Most of the
change in residual variance indicated in these results was recorded at school level
rather than teacher level. Multilevel Model 6 (Table 5.8) indicates the addition of
school average level school learning environment explanatory variables. One
163
variable is indicated as having a statistically significant, positive influence on
teachers’ global satisfaction with leadership scores, professional interest (PIs).
Reduced Model 6 (Table 5.8) indicates the result of removing all non-
significant explanatory variables in accounting for variations in teachers’ innovation
scores. Four explanatory variables were indicated as accounting for 35.5% of the
variance in teachers’ global satisfaction with leadership scores. Three variables were
indicated to have a statistically significant, negative influence on teachers’ global
satisfaction with leadership scores: the teacher level demographic variable teacher
position (Teacht), teacher level teacher current time in school (Tcurt) and school
average level laissez-faire leadership (LFs). Predictably, teachers’ perception of
global satisfaction with leadership (SAT) increased with a fall in laissez-faire (LFs)
leadership style behaviour. These results also suggest that younger, inexperienced
teachers are more satisfied, and perhaps less cynical than older teachers who are
more experienced in the profession. One explanatory variable was indicated as
having a statistically significant, positive influence on global satisfaction scores,
namely, school-average level individualised consideration (ICs). The analysis
suggests that teachers in schools respond favourably to the individualised
consideration (ICs) behaviours of their principals. This finding is somewhat
controversial in that it is contrary to transformational leadership literature describing
the role vision plays both in obtaining extra effort from followers and as a motivator
that induces follower satisfaction (Bass et al., 1997).
It was expected that the leadership style of principals would influence
teachers’ perception of global satisfaction with leadership (SAT) (Hypothesis 2). It
was further expected, according to the transformational leadership literature (Bass et
al., 1997), that visionary leadership behaviour (VIs) would demonstrate a marked
influence on leadership satisfaction. However, the results of this analysis have
indicated that teachers’ perception of leadership satisfaction is more positively
influenced by a principal’s individualised consideration (ICs) behaviour than it is by
visionary (VIs) behaviour.
Teachers’ Perception of Influence. Table 5.9 below shows the result of
fitting multilevel regression models to teachers’ perception of influence (INFLU)
164
scores by adding four blocks of explanatory variables. Teachers’ perception of
influence is defined as the perception among teachers of the amount of influence they
exert in their schools, for example, in the area of decision making.
Initially a two level variance components model was fitted to teachers’
perception of influence (INFLU) scores, the result of which is shown as Multilevel
Model 1 (Table 5.9). Only the variance recorded at teacher level is indicated as being
statistically significant. Figure 5.16 shows the ranked residuals plotted for each of the
school level teachers’ perception of influence (INFLU) data sets. Only one school is
indicated as departing significantly from the overall average line as predicted by the
fixed parameter, and is indicated as lying above the average line.
Figure 5.16. Residuals of school level teachers’ perception of influence (INFLU)
scores, ranked for 52 schools, showing 95% confidence intervals.
Multilevel Model 2 (Table 5.9) shows the results of fitting only the three
leadership variables as explanatory variables, namely vision (VIs), individualised
consideration (ICs) and laissez-faire leadership (LFs). None of three leadership
variables had a statistically significant effect on teachers’ perception of influence
(INFLU) scores.
165
Table 5.9
Variation in Teachers’ Perception of Influence Scores (N = 458) in 52 Schools Explanatory Variables
Multilevel Model 1
Multilevel Model 2
Multilevel Model 3
Multilevel Model 4
Multilevel Model 5
Multilevel Model 6
Reduced Model 6
Fixed:
Constant
(β0jX0): -.018(.052) -.014(.049) -.013(.048) -.013(.047) -.012(.045) -.012(.043) -.012(.043)
School Level Demographic
X1 Teachs (β1) .004(.051) -.020(.051) -.042(.050) .007(.056)
X2 Texps (β2 ) .106(.084) .127(.092) .121(.089) .218(.111)* .177(.076)*
X3 Ssizs (β3) -.005(.050) -.006(.049) -.053(.053) -.036(.054)
X4 Tages (β4) -.156(.081) -.136(.086) -.128(.084) -.206(.093)* -.189(.074)*
X5 Tgens (β5) -.059(.050) -.009(.053) -.030(.052) .043(.064)
X6 Pgens (β6) .050(.049) .055(.050) -.011(.051) -.016(.060)
Teacher Level Demographic
X7 Teacht (β7) .094(.047)* .092(.047)* .100(.047)* .128(.044)*
X8 Texpt (β8 ) -.099(.080) -.101(.080) -.102(.079)
X9 Tcurt (β9) .058(.057) .062(.057) .052(.056)
X10 Taget (β10) -.022(.073) -.023(.073) -.019(.072)
X11 Tgent (β11) -.134(.049)* -.133(.048)* -.135(.048)* -.118(.045)*
X12TSCPt (β12) .042(.050) .044(.051) .068(.053)
School Level Leadership
X13 VIs (β13) .009(.078) .070(.081) .043(.087)
X14 ICs (β14 ) -.144(.077) -.180(.087)* -.170(.093) -.160(.058)*
X15 LFs (β15) -.131(.068) -.130(.067) -.162(.096) -.148(.058)*
School Level School Learning Environment
X16 SSs (β18) -.120(.064) -.113(.050)*
X17 AFFs (β19) .018(.061)
X18 PIs (β18) -.056(.075)
X19 CENs (β19) .003(.079)
X20 INNs (β20) -.019(.089)
X21 RAs (β21) -.114(.063) -.117(.048)*
X22 AOs (β22) .032(.066)
Random:
u0j (School-level) .033(.027) .020(.024) .013(.023) .015(.022) .004(.020) .000(.000) .000(.000)
e0ij (Teacher-level) .902(.063)* .902(.063)* .903(.063)* .871(.061)* .870(.061)* .843(.056)* .856(.057)*
% of Variance
Explained 1.4% 2.0% 5.2% 6.5% 9.8% 8.4%
-2 (Log-Likelihood) 1266.677 1261.324 1259.378 1243.604 1238.129 1221.596 1228.721
Note. IGLS solutions shown; fitted parameter estimates with standard errors in parentheses. * Statistically significant beyond the p < .05 level by univariate two tailed test: i.e., the
parameter estimate is greater than twice its standard error.
166
Multilevel Model 3 (Table 5.9) indicates the amount of variance that can be
attributed with the addition of a block of school level demographic explanatory
variables. None of these six school level demographic variables are indicated as
being statistically significant. This is not surprising, given that there is almost no
school level variance to explain. Multilevel Model 4 (Table 5.9) indicates the
influence of six teacher level demographic explanatory variables on teachers’
perception of influence (INFLU) scores. Two variables were indicated as influencing
teachers’ perception of influence scores. Teacher position (Teacht) was indicated as
having a statistically significant, positive influence on teachers’ perception of
influence (INFLU) scores, while teacher gender (Tgent) was demonstrated as having
a statistically significant, negative influence on teachers’ influence scores (INFLU).
This result suggests that teachers perceive that executive staff have more influence
over workload than do classroom teachers, and that male teachers have more
influence over their workload than do female teachers.
Multilevel Model 5 (Table 5.9) adds in the cumulative influence of three
leadership style behaviours as explanatory variables in accounting for variance in
teachers’ perception of influence (INFLU) scores. Only school average level
individualised consideration (ICs) was indicated as having a statistically significant,
negative influence on these teacher perception scores. Teachers’ perception is that in
schools where principal’s individualised consideration (ICs) behaviour increases,
there is a corresponding decrease in perception of teacher influence (INFLU). This
result may indicate that “insider” staff, namely, those close to the principal, are the
ones with influence, and that most teachers are not part of this “insiders” group.
While this result seems curious and may well be contrary to what might be expected,
it needs further investigation.
Individual teacher gender (Tgent) also indicated a negative, statistically
significant relation to teachers’ perception of influence (INFLU). This indicates that
male teachers’ perception is that they exert more influence than female teachers do.
Teacher role (Teacht) is indicated as having a statistically significant, positive effect
on teachers’ perception of influence (INFLU) scores, suggesting that holding an
executive position brings opportunity for increased influence. This signals that male
167
teachers perceive that they exert more influence over their workloads than do female
teachers.
The addition of seven school average level school learning environment
explanatory variables is indicated in Multilevel Model 6 (Table 5.9). No one school
learning environment factor was indicated as having any statistically significant
influence of teachers’ perception of influence scores. The addition of this school
learning environment block of explanatory variables, along with the other three
blocks of explanatory variables, accounted for 9.8% of the recorded variance.
Reduced Multilevel Model 6 (Table 5.9) indicated the effect of removing all
non-significant explanatory variables from the model. In total, eight explanatory
variables were indicated as statistically significant, and accounted for 8.4% of the
variance. Six explanatory variables were indicated as having a statistically
significant, negative influence on teachers’ perceptions of influence. They included,
at the school-average level, teacher age (Tages), individualised consideration (ICs),
laissez-faire leadership (LFs), student supportiveness (SSs), resource adequacy
(RAs), and, at the teacher level, teacher gender (Tgent). At the school-average level,
these results suggest that schools that have a higher proportion of younger teachers,
schools where female principals led, schools where teachers perceived the leadership
style of principals to be individualised or laissez-faire, schools where student
supportiveness was low and where resources were few, were factors that increased
teachers’ perception of influence in their respective schools. At the individual teacher
level, male teachers perceived they exerted more influence in their schools than
female teachers did.
Two explanatory variables were indicated as having a statistically significant,
positive influence on these teachers’ perception of influence (INFLU) scores: at the
school level, teacher experience (Texps), and at the teacher level teacher position
(Teacht). Teachers’ perception was that the higher the school-average level of
experience, the more influence they exerted in their schools. Further, within schools,
head teachers exerted more influence than did classroom teachers.
168
Two leadership explanatory variables, individualised consideration (ICs) and
laissez-faire (LFs) leadership, indicated a statistically significant, negative influence
on teachers’ perception of influence (INFLU) scores (Hypothesis 2.2). These results
are contrary to predictions made in the transformational leadership literature. Further,
several school and teacher level demographic variables were also indicated as having
statistically significant influences on these teachers’ influence scores (Research
Question 3).
Teachers’ Perception of Effectiveness. Table 5.10 shows the result of fitting
multilevel regression models to teachers’ perception of effectiveness (EFF) scores by
adding four blocks of explanatory variables. Teachers’ perception of effectiveness
can be defined as teachers’ ability to meet the educational needs of their students.
A two level variance components model was fitted to teachers’ perception of
effectiveness (EFF) scores, the result of which is shown as Multilevel Model 1
(Table 5.10). Variance was recorded at teacher level and at the school level, both of
which were indicated as being statistically significant. Figure 5.17 shows the ranked
residuals plotted for each of the school level teachers’ perception of effectiveness
(EFF) data sets. Several schools are indicated as departing significantly from the
overall average line, and are indicated as lying both above and below the average
line. Hence, this analysis indicated substantial school-to-school differences.
169
Figure 5.17. Residuals of school level teachers’ perception of effectiveness (EFF)
scores, ranked for 52 schools, showing 95% confidence intervals.
Multilevel Model 2 (Table 5.10) shows the effect of using the three
leadership explanatory variables only to account for the variance recorded in
teachers’ perception of effectiveness (EFF) scores. None of the three leadership
variables, vision (VIs), individualised consideration (ICs) and laissez-faire leadership
(LFs), were indicated as having a statistically significant influence on these teacher
perception scores. The recorded change in the log likelihood statistic also indicates
that the addition of this block of explanatory variables was not significant.
170
Table 5.10
Variation in Teachers’ Perception of Effectiveness Scores (N = 458) in 52 Schools Explanatory Variables
Multilevel Model 1
Multilevel Model 2
Multilevel Model 3
Multilevel Model 4
Multilevel Model 5
Multilevel Model 6
Reduced Model 6
Fixed:
Constant
(β0jX0): -.026(.079) -.020(.076) -.016(.075) -.016(.074) -.013(.072) -.008(.063) -.014(.068)
School Level Demographic
X1 Teachs (β1) -.127(.076) -.130(.076) -.125(.075) -.098(.077)
X2 Texps (β2 ) .094(.116) .076(.120) .059(.118) -.066(.138)
X3 Ssizs (β3) .008(.074) .009(.074) -.002(.081) .012(.075)
X4 Tages (β4) -.189(.112) -.192(.115) -.164(.114) -.036(.119)
X5 Tgens (β5) .029(.076) .029(.077) .009(.077) -.135(.086)
X6 Pgens (β6) .009(.077) .005(.077) -.001(.081) -.101(.088)
Teacher Level Demographic
X7 Teacht (β7) -.002(.045) .000(.045) -.002(.044)
X8 Texpt (β8 ) .012(.073) .008(.074) .013(.073)
X9 Tcurt (β9) -.009(.055) -.010(.055) .006(.055)
X10 Taget (β10) .023(.067) .025(.067) .020(.067)
X11 Tgent (β11) .009(.045) .008(.045) .019(.067)
X12TSCPt (β12) .057(.056) .077(.057) .031(.057)
School Level Leadership
X13 VIs (β13) .168(.117) .163(.124) .082(.119)
X14 ICs (β14 ) -.010(.116) -.007(.136) -.174(.133)
X15 LFs (β15) .033(.103) .046(.107) -.072(.130)
School Level School Learning Environment
X16 SSs (β16) .059(.086)
X17 AFFs (β17) .041(.085)
X18 PIs (β18) .071(.103)
X19 CENs (β19) .081(.106)
X20 INNs (β20) .224(.120)* .287(.068)*
X21 RAs (β21) -.109(.087)
X22 AOs (β22) .038(.083)
Random:
u0j (School-level) .223(.062)* .201(.058) .192(.056)* .188(.055)* .169(.051)* .110(.039)* .142(.046)*
e0ij (Teacher-level) .736(.052)* .736(.052) .734(.051)* .732.051)* .733(.051)* .731(.051)* .736(.052)*
% of Variance
Explained 2.3% 3.4% 4.1% 5.9% 12.3% 8.4%
-2 (Log-Likelihood) 1224.255 1220.704 1217.915 1216.156 1213.068 1198.612 1208.977
Note. IGLS solutions shown; fitted parameter estimates with standard errors in parentheses. * Statistically significant beyond the p < .05 level by univariate two tailed test: i.e., the
parameter estimate is greater than twice its standard error.
171
Multilevel Models 3, 4 and 5 (Table 5.10) indicate the accumulative addition
of three blocks of explanatory variables to explain the variance recorded in teachers’
perception of effectiveness (EFF) scores. While the addition of each of the three
blocks of explanatory variables accounted for an increasing amount of the variance
recorded (school-average level demographic variables, 3.4%; teacher level
demographic variables, 0.7%; and leadership variables, 1.8%), none of the variables
was statistically significant. Further, the changes in the log likelihood statistic
between Multilevel Models 3, 4 and 5 (Table 5.10) also indicate that the addition of
these blocks of explanatory variables was not statistically significant.
Seven school average level school learning environment explanatory
variables are added in Multilevel Model 6 (Table 5.10). Initially, only one factor,
innovation (INNs) was indicated as having a positive, statistically significant
influence on teachers’ perception of effectiveness scores. This additional block of
school level school learning environment explanatory variables, along with the other
three blocks of variables, accounted for 12.3% of the recorded variance. The random
part of the model indicates records that residual variance is statistically significant at
both teacher and school levels, suggesting that the explanatory variables cannot
account for all the variance recorded. Reduced Multilevel Model 6 (Table 5.10)
indicated the effect of removing all non-significant explanatory variables from the
model. Only one explanatory variable, innovation (INNs) was indicated as being
statistically significant in accounting for the variation in teachers’ perception of
effectiveness scores. It alone accounted for 8.4% of the recorded variation in these
scores. The change in the school level parameter estimate between Multilevel Model
1 and Reduced Model 6 indicates that all of the variance was found at the school
level. Both school and teacher level residual variation was indicated as being
statistically significant. None of the three leadership explanatory variables, vision
(VIs), individualised consideration (ICs) and laissez-faire leadership (LFs) was
indicated as having a statistically significant influence on teachers’ perception of
effectiveness scores.
None of the three leadership variables under consideration, namely vision
(VIs), individualised consideration (ICs) or laissez-faire leadership (LFs) indicated
any statistically significant influence on teachers’ perception of effectiveness (EFF)
172
scores (Hypothesis 2). Furthermore, none of the school level or teacher level
antecedent variables demonstrated any statistically significant influence on teachers’
perception of effectiveness (EFF) scores (Research Question 3).
Teachers’ Perception of Control. Table 5.11 shows the result of fitting
multilevel regression models to teachers’ perception of control (TCON) scores.
Teachers’ perception of control can be defined as the degree to which teachers
perceived they control their working conditions and workloads, as opposed to being
controlled by executive staff.
A two level variance components model was fitted to teachers’ perception of
control (TCON) scores, the result of which is shown as Multilevel Model 1 (Table
5.11). While variance was recorded at teacher level and at the school level, only
variation at the teacher level was indicated as being statistically significant.
Figure 5.18 shows the ranked residuals plotted for each of the school level
teachers’ perception of control (TCON) data sets. One school was indicated as
departing significantly from the overall average line, as predicted by the fixed
parameter, and is indicated as lying below the average line.
173
Table 5.11
Variation in Teachers’ Perception of Control Scores (N = 458) in 52 Schools
Explanatory Variables
Multilevel Model 1
MultilevelModel 2
MultilevelModel 3
Multilevel Model 4
Multilevel Model 5
Multilevel Model 6
Reduced Model 6
Fixed:
Constant
(β0jX0): -.007(.055) -.008(.053) -.007(.054) -.008(.053) -.008(.050) -.005(.044) -.005(.045)
School Level Demographic
X1 Teachs (β1) .045(.057) .071(.056) .095(.055) .047(.057)
X2 Texps (β2 ) .067(.093) .029(.099) .029(.096) .072(.113)
X3 Ssizs (β3) -.030(.055) -.030(.054) -.009(.059) .039(.054)
X4 Tages (β4) -.037(.089) .008(.093) .009(.091) -.014(.095)
X5 Tgens (β5) .039(.056) .034(.058) .043(.057) -.005(.065)
X6 Pgens (β6) .005(.056) -.006(.055) .036(.057) .088(.061)
Teacher Level Demographic
X7 Teacht (β7) -.109(.048)* -.106(.048)* -.105(.048)* -.124(.045)*
X8 Texpt (β8 ) .143(.081) .140(.081) .148(.080)
X9 Tcurt (β9) -.101(.059) -.101(.058) -.097(.057)
X10 Taget (β10) -.116(.074) -.114(.074) -.117(.073)
X11 Tgent (β11) -.018(.049) -.019(.049) -.018(.049)
X12TSCPt (β12) -.018(.053) -.001(.053) -.041(.054)
School Level Leadership
X13 VIs (β13) .077(.084) .032(.090) -.067(.089)
X14 ICs (β14 ) .116(.082) .171(.097) .038(.094)
X15 LFs (β15) .129(.073) .165(.076)* .045(.097)
School Level School Learning Environment
X16 SSs (β16) -.040(.065)
X17 AFFs (β17) .223(.062)* .211(.045)*
X18 PIs (β18) -.057(.076)
X19 CENs (β19) .140(.080)
X20 INNs (β20) .024(.090)
X21 RAs (β21) .044(.064)
X22 AOs (β22) -.035(.067)
Random:
u0j (School-level) .047(.030) .035(.028) .040(.029) .037(.028) .027(.025) .000(.000) .001(.020)
e0ij (Teacher-level) .922(.064)* .918(.064)* .923(.064)* .895(.062)* .890(.062)* .872(.058)* .910(.063)*
% of Variance
Explained 1.7% .6% 3.8% 5.4% 10.0% 6.0%
-2 (Log-Likelihood) 1281.406 1275.344 1279.365 1265.012 1258.265 1237.053 1256.894
Note. IGLS solutions shown; fitted parameter estimates with standard errors in parentheses. * Statistically significant beyond the p < .05 level by univariate two tailed test: i.e., the
parameter estimate is greater than twice its standard error.
174
Figure 5.18. Residuals of school level teachers’ perception of control (TCON)
scores, ranked for 52 schools, showing 95% confidence intervals.
Multilevel Model 2 (Table 5.11) shows the result of fitting three leadership
explanatory variables, vision (VIs), individualised consideration (ICs) and laissez-
faire leadership (LFs), to account for the variation in teachers’ perception of control
scores. None of these three leadership variables was indicated as having any
statistically significant influence on the teachers’ perception of control scores. The
small change in the log-likelihood statistic also indicates that the effect of these three
leadership explanatory variables is non-significant.
Multilevel Model 3 (Table 5.11) includes the influence of a block of school-
average level demographic variables to account for the variation in teachers’
perception of control scores. None of these six explanatory variables was indicated as
having any statistically significant influence on these teacher perception scores.
Multilevel Model 4 (Table 5.11) includes a block of six teacher level explanatory
demographic variables along with the block of school-average level demographic
variables examined in Multilevel Model 3 (Table 5.11). One explanatory variable,
teacher position (Teacht) in the school (i.e., whether administrative or teaching staff),
was indicated as being statistically significant. These two blocks of explanatory
175
variables accounted for 3.8% of the recorded variance. Multilevel Model 5 (Table
5.11) added the three leadership explanatory variables, vision (VIs), individualised
consideration (ICs) and laissez-faire leadership (LFs). Laissez-faire leadership (LFs)
was initially indicated as being statistically significant, along with teacher level
teacher position (Teacht). Together, these three blocks of explanatory variables
accounted for 5.4% of the recorded variance in teachers’ perception of control
scores.
Multilevel Model 6 (Table 5.11) included the influence of school average
school
was unexpected that teachers’ perception of control (TCON) scores were
demon
learning environment explanatory variables to account for the variation in
teachers’ perception of control scores. Of the seven school learning environment
variables considered, only affiliation (AFFs) was indicated as being statistically
significant, with the four blocks of explanatory variables together accounting for
10.0% of the recorded variance. Reduced Model 6 (Table 5.11) indicated the effect
of removing all non-significant explanatory variables from the model. Two
explanatory variables were indicated as statistically significant, and accounted for
6.0% of the variance, that variance recorded at both the school and the teacher level.
One explanatory variable was indicated as having a statistically significant, negative
influence on teachers’ perception of control scores, teacher level teacher position
(Teacht). One explanatory variable, affiliation (AFFs) was indicated as having a
statistically significant, positive influence on teachers’ perception of control scores.
These results indicate that higher perceptions of control are associated with higher
school-average perceptions of affiliation, and that teaching staff, as opposed to
administrative staff, have a lower perception of control.
It
strated not to be statistically significantly influenced by any of the three
leadership explanatory variables considered in this analysis (Hypotheses 2.1 and 2.2).
Further, only two variables, teacher level teacher position (Teacht) and school
average level affiliation (AFFs) indicated any statistically significant influence on
teachers’ perception of control (TCON) scores.
176
Hypothesis 2.1: Results
has been hypothesised that teachers who have experienced a
transfo
he transformational leadership behaviour of visionary leadership (VIs) was
indicat
he absence of any statistically significant score for visionary leadership
(VIs) a
Hypothesis 2.2: Results
he results indicate that principal’s individualised consideration (ICs)
behavio
It
rmational leadership style will display higher scores compared to teachers
who have experienced transactional or laissez-faire leadership styles for satisfaction
with their leader, and ratings of their effectiveness, influence and control as teachers.
A series of four multilevel models, shown in Tables 5.8, 5.9, 5.10 and 5.11, was
employed to test this hypothesis.
T
ed as non-significant in each of the four teacher outcome measures of global
satisfaction with leadership (SAT; Table 5.8), perceptions of influence (INFLU;
Table 5.9), perceptions of effectiveness (EFF; Table 5.10) and perceptions of control
(TCON; Table 5.11). This suggests that principal’s visionary style leadership
behaviour had no effect on these four teacher outcome measures.
T
s a possible explanatory variable in accounting for variance in each of the four
teacher outcome measures indicates that Hypothesis 2.1 has to be rejected.
T
ur has a strong, statistically significant, positive influence on teachers’
perception of global satisfaction with leadership (SAT; Table 5.8). Furthermore, the
analysis also indicated that the influence of principal’s individualised consideration
(ICs) behaviour is greater than the influence of principals’ laissez-faire leadership
(LFs) behaviour, which although statistically significant, is negative in direction.
This indicates that global satisfaction with leadership (SAT) was more responsive to
individualised consideration (ICs) as a leadership style behaviour than it was to the
other leadership behaviours, laissez-faire leadership (LFs) and vision (VIs).
Therefore, in the case of teachers’ perception of global satisfaction with leadership
(SAT), Hypothesis 2.2 is accepted.
177
The teacher outcome of teachers’ perception of influence (INFLU; Table 5.9)
is like
he teacher outcomes of teachers’ perception of effectiveness (EFF; Table
5.10) a
Hypothesis 2.1. and 2.2: Summary
ased on the analysis presented, Hypothesis 2.1 is rejected. Hypothesis 2.2 is
accepte
n analysis of the data gathered from 458 teachers across 52 secondary
schools
owever, the analysis did indicate that principal’s individualised
conside
wise indicated as being more responsive to the leadership behaviour of
individualised consideration (ICs), although it is recorded in a negative direction,
than it is to the leadership behaviour laissez-faire leadership (LFs). Therefore, in the
case of teachers’ perception of influence (INFLU), Hypothesis 2.2 is rejected.
T
nd teachers’ perception of control (TCON; Table 5.11) are indicated as
showing no statistically significant responsiveness to any of the three leadership style
behaviours under consideration. Therefore, for teachers’ perception of effectiveness
(EFF) and teachers’ perception of control (TCON), Hypothesis 2.2 is rejected.
B
d in the case of global satisfaction with leadership, and rejected in the cases
of perceptions of influence, effectiveness and control.
A
in New South Wales indicated that a principal’s visionary behaviour had no
statistically significant influence on teachers’ perceptions of four teacher outcomes.
As in Hypothesis 1.1 and 1.2, this is contrary to the transformational leadership
literature, which suggests that transformational leadership behaviour is more likely to
motivate followers to higher levels of satisfaction and performance than are
transactional or laissez-faire leadership behaviours.
H
ration (ICs) behaviours were more influential than principal’s visionary (VIs)
behaviours in influencing teachers’ perception of global satisfaction with leadership
scores. Furthermore, and consistent with a priori predictions, teachers indicated that
principal’s laissez-faire (LFs) leadership behaviours impacted global satisfaction
with leadership and perceptions of influence to a statistically significant but negative
effect.
178
Results: Research Question 3. What Relation do Antecedent
Background Variables have with School Leadership, School
Learning Environment and Teacher Outcomes?
Overview
It has been predicted that the antecedent background variables considered in
this study will demonstrate differential impact in terms of teacher perceptions of
leadership style, learning environment and teacher outcomes. These antecedent
variables include teachers’ position on staff (administrative or non-administrative),
the amount of time a teacher has been teaching, the amount of time a teacher has
spent in a school, the size of a school’s student population, teacher’s age, teacher’s
gender, principal’s gender and the amount of time a teacher has spent working with
their current principal. Further, even though it could be argued that principals can
influence many of these antecedent variables in terms of how staff is selected, it is
necessary to consider the effects of these antecedent background variables as a pre-
condition to understanding the effects of the other variables in this analysis
(Hallinger & Heck, 1998). A series of multilevel models was developed to analyse
the relation between these antecedent demographic background variables, both at
school and teacher level, and the response variables of school leadership, school
learning environment and teacher outcomes.
In the following analyses of vision, individualised consideration and laissez-
faire leadership, individual teacher ratings of these three leadership styles were
considered and not school-average level data. Therefore, unlike the previous
analysis, where vision, individualised consideration and laissez-faire leadership were
indicated at the school average level with the subscript s, they are indicated as
individual teacher level scores with the subscript t.
Analysis
Multilevel modelling allows the effect of various explanatory variables on
response variables to be considered within a context of the nested structure of much
179
educational data (Rowe, 1995; Goldstein et al., 1998), including the data gathered in
this study. In this section, three multilevel models were used to examine the relation
between school and teacher level demographic explanatory variables and the
leadership response variables of vision (VIs), individualised consideration (ICs) and
laissez-faire leadership (LFs). The results of these analyses are shown below in
Tables 5.12, 5.13 and 5.14 respectively.
The Effect of Antecedent Variables on School Leadership
Vision. Multilevel Model 1 (Table 5.12) indicates the results of fitting a two-
level variance components model to teachers’ perception of principal’s vision (VIt)
scores. The random part of the model indicated that residual variance at both the
school and teacher level was found to be statistically significant. Multilevel Model 2
(Table 5.12) indicates the amount of variance that can be attributed to the six school
average level demographic explanatory variables. No school level demographic
explanatory variable was indicated to be statistically significant in accounting for the
variance recorded in teachers’ perception of principal’s vision (VIt) scores.
Multilevel Model 3 (Table 5.12) indicates the amount of variance that can be
attributed to the six teacher level demographic explanatory variables. Three were
demonstrated to be statistically significant in their effect on teachers’ perception of
principals’ vision (VIt) scores. Teacher position (Teacht) and teacher time with
current principal (TSCPt) were both indicated as having a statistically significant,
negative influence on teachers’ vision scores, while teacher gender (Tgent) was
demonstrated as having a positive effect on teacher’s scores. Reduced Multilevel
Model 3 (Table 5.12) indicates the result of removing all non-significant explanatory
variables from the model. Three explanatory variables, teacher position (Teacht),
teacher time with current principal (TSCPt), and teacher gender (Tgent) were
indicated as remaining in the model, and accounting for 10.3% of the recorded
variance in teachers’ perception of principal’s vision scores. The teacher level
variable of individual teacher gender (Tgent) is shown as having a statistically
significant, positive influence on teachers’ perception of principals’ vision (VIt)
scores, indicating that individual male and female teachers respond differently to
their principal’s vision (VIt). Two teacher level explanatory variables, teacher
position in school (Teacht) and time spent in school with their current principal
180
(TSCPt) are shown as having a statistically significant, negative influence on
teacher’s perception of principal’s vision scores. This suggests that individual staff
who are in executive positions respond differently to their principal’s vision (VIt), as
do teachers who have served in their schools with their current principals for longer
time periods.
Table 5.12
Variation in Teachers’ Perception of Principal’s Vision Scores (N = 458) in 52 Schools Explanatory Variables
MultilevelModel 1
MultilevelModel 2
MultilevelModel 3
Reduced Multilevel Model 3
Fixed:
Constant
(β0jX0): -.027(.081) -.022(.077) -.025(.071) -.033(.074)
School Level Demographics
X1 Teachs (β1) -.037(.079) .010(.073)
X2 Texps (β2 ) .042(.118) .085(.115)
X3 Ssizs (β3) .006(.076) .005(.070)
X4 Tages (β4) -.116(.114) -.123(.110)
X5 Tgens (β5) .108(.077) .056(.073)
X6 Pgens (β6) .077(.080) .072(.074)
Teacher Level Demographics
X7 Teacht (β7) -.181(.043)* -.181(.041)*
X8 Texpt (β8 ) -.007(.070)
X9 Tcurt (β9) -.064(.053)
X10 Taget (β10) -.006(.064)
X11 Tgent (β11) .110(.042)* .125(.041)*
X12 TSCPt (β12) -.144(.054)* -.173(.049)*
Random:
u0j (School-level) .242(.065)* .210(.059)* .172(.050)* .193(.055)*
e0ij (Teacher-level) .721(.050)* .720(.050)* .668(.047)* .671(.047)*
% of Variance
Explained 3.4% 12.8% 10.3%
-2 (Log-Likelihood) 1218.309 1212.984 1173.954 1180.003
Note. IGLS solutions shown; fitted parameter estimates with standard errors in parentheses. * Statistically significant beyond the p < .05 level by univariate two tailed test: i.e., the
parameter estimate is greater than twice its standard error.
Two teacher level antecedent variables, teacher position in school (Teacht)
and time spent with current principal (TSCPt) demonstrated a statistically significant,
negative influence on teachers’ perception of principals’ vision (VIt) scores, while
181
teacher gender (Tgent) indicated a statistically significant, positive influence on these
teachers’ scores (Research Question 3).
Individualised Consideration. Table 5.13 below indicates the results of
fitting multilevel regression models to teachers’ perception of principals’
individualised consideration (ICt) scores, using school average and teacher level
demographic scores as explanatory variables to account for any recorded variance.
The leadership response variable of teachers’ perception of individualised
consideration (ICt) scores is an individual teacher level variable.
Multilevel Model 1 (Table 5.13) indicates the results of fitting a two-level
variance components model to teachers’ perception of principals’ individualised
consideration (ICt) scores. The random part of the model indicates that residual
variance was found to be statistically significant. The addition of school average level
demographic explanatory variables is indicated in Multilevel Model 2 (Table 5.13).
Only one school level demographic variable, school size (Ssizs) is indicated as being
statistically significant, having a negative influence on teachers’ perception of
principals’ individualised consideration (ICt) scores. However, a non-significant
decrease in the log-likelihood statistic indicates that these six school average
demographic explanatory variables were not influential in affecting teachers’
individualised consideration (ICt) scores. Multilevel Model 3 (Table 5.13) adds in the
effect of six teacher level demographic explanatory variables. One teacher-level
explanatory variable, teacher position (Teacht) along with school level school size
(Ssizs), was indicated as having a statistically significant, negative effect on teachers’
perception of principals’ individualised consideration (ICt) scores. The random part
of the model indicated that residual variance was significant.
Reduced Multilevel Model 3 (Table 5.13) indicates the results of removing all
non-significant explanatory variables from the model. Three explanatory variables,
school level school size (Ssizs), teacher level teacher position (Teacht) and teacher
length of time in their current school (Tcurt) all exerted a statistically significant,
negative influence on teachers’ perceptions of principals’ individualised consideration
(ICt) scores. Together, these explanatory variables accounted for 10.7% of the
recorded variance. It was interpreted that, at the school level, teachers perceive that as
182
schools get larger in terms of the numbers of student within the school (Ssizs),
principals’ individualised consideration (ICt) behaviours become less effective. This
is an expected effect as schools increase in their student and teacher populations. At
the teacher level, however, teachers’ perceive that executive teachers (Teacht) are less
influenced by their principal’s individualised consideration (ICt) behaviour than are
teachers who have served long periods of time in their schools (Tcurt). While it is
difficult to understand this relation, this may be a by-product of familiarity with the
principal’s leadership style as a result of working closely with them, or it may signal
the effect of teacher recognition that principals are answerable to head offices in the
implementation of unpopular policies.
Table 5.13 Variation in Teachers’ Perception of Principal’s Individualised Consideration Scores (N = 458) in 52 Schools Explanatory Variables
MultilevelModel 1
MultilevelModel 2
MultilevelModel 3
Reduced Multilevel Model 3
Fixed:
Constant
(β0jX0): .000(.075) -.009(.071) -.013(.068) -.015(.068)
School Level Demographics
X1 Teachs (β1) -.081(.073) -.030(.070)
X2 Texps (β2 ) .021(.113) .071(.112)
X3 Ssizs (β3) -.163(.071)* -.163(.067)* -.157(.064)*
X4 Tages (β4) -.081(.109) -.066(.107)
X5 Tgens (β5) .022(.072) -.005(.070)
X6 Pgens (β6) .002(.074) .007(.070)
Teacher Level Demographics
X7 Teacht (β7) -.215(.044)* -.200(.041)*
X8 Texpt (β8 ) -.017(.073)
X9 Tcurt (β9) -.095(.054) -.166(.043)*
X10 Taget (β10) -.052(.067)
X11 Tgent (β11) .029(.044)
X12 TSCPt (β12) -.104(.054)
Random:
u0j (School-level) .193(.057)* .161(.051)* .142(.045)* .146(.046)*
e0ij (Teacher-level) .787(.055)* .785(.055)* .719(.050)* .729(.051)*
% of Variance
Explained 3.5% 12.1% 10.7%
-2 (Log-Likelihood) 1247.486 1240.475 1199.031 1205.727
Note. IGLS solutions shown; fitted parameter estimates with standard errors in parentheses. * Statistically significant beyond the p < .05 level by univariate two tailed test: i.e., the
parameter estimate is greater than twice its standard error.
183
Three antecedent variables indicated a negative, statistically significant
influence on teachers’ perception of principals’ individualised consideration (ICt)
scores (Research Question 3). At the school level, school size (Ssizs) in terms of
student numbers influenced teachers’ perception scores. At teacher level, teacher
position (Teacht) and teacher length of time served in their current school (Tcurt)
were also indicated as having a statistically significant, negative influence on
teachers’ perception of principals’ individualised consideration (ICt) scores.
Laissez-Faire Leadership. Table 5.14 indicates the results of fitting
multilevel regression models to individual teachers’ perceptions of principals’ laissez-
faire leadership (LFt) scores.
Multilevel Model 1 (Table 5.14) indicates the results of fitting a two-level
variance components model to teachers’ perceptions of principals’ laissez-faire
leadership (LFt) scores. The random part of the model indicates that variance was
found to be statistically significant at both the individual teacher and school levels.
Multilevel Model 2 (Table 5.14) indicates the result of fitting school level
demographic explanatory variables to account for variance recorded in teachers’
perceptions of principals’ laissez-faire leadership (LFt) scores. Only principal gender
(Pgens) at the school level was indicated as having a statistically significant, negative
influence on teachers’ laissez-faire leadership scores. Multilevel Model 3 (Table 5.14)
shows the result of fitting six teacher level demographic explanatory variables to
account for variance recorded in teachers’ perceptions of principals’ laissez-faire
leadership (LFt) scores. One variable, teacher length of time in their current school
(Tcurt), was indicated as having a statistically significant, positive effect on teachers’
laissez-faire leadership (LFt) scores. A second variable, principal gender (Pgens) at
the school level was indicated as having a statistically significant negative effect on
teachers’ perceptions of principals’ laissez-faire leadership (LFt) scores. Reduced
Multilevel Model 3 (Table 5.14) indicates the result of removing all non-significant
explanatory variables from the analysis. Only teacher length of time in their current
school (Tcurt) remained as indicating any statistical significance, accounting for 5.7%
in the variation recorded in teachers’ scores. The statistical significance of the
residual variance parameters indicates that the explanatory variables failed to account
for all the recorded variance in this model.
184
Table 5.14
Variation in Teachers’ Perceptions of Principal’s Laissez-Faire Leadership Scores (N = 458) in 52 Schools Explanatory Variables
MultilevelModel 1
MultilevelModel 2
MultilevelModel 3
Reduced Multilevel Model 3
Fixed:
Constant
(β0jX0): .036(.076) .033(.072) .035(.070) .037(.073)
School Level Demographics
X1 Teachs (β1) -.027(.073) -.051(.072)
X2 Texps (β2 ) -.043(.113) -.165(.115)
X3 Ssizs (β3) .045(.071) .048(.069)
X4 Tages (β4) .111(.109) .145(.110)
X5 Tgens (β5) -.097(.073) -.057(.072)
X6 Pgens (β6) -.157(.074)* -.145(.073)*
Teacher Level Demographics
X7 Teacht (β7) .066(.043)
X8 Texpt (β8 ) .138(.072)
X9 Tcurt (β9) .131(.053)* .212(.043)*
X10 Taget (β10) -.053(.066)
X11 Tgent (β11) -.063(.043)
X12 TSCPt (β12) .075(.054)
Random:
u0j (School-level) .198(.057)* .168(.051)* .161(.049)* .179(.053)*
e0ij (Teacher-level) .752(.053)* .747(.052)* .695(.049)* .717(.050)*
% of Variance
Explained 3.7% 9.9% 5.7%
-2 (Log-Likelihood) 1228.995 1221.018 1188.913 1205.315
Note. IGLS solutions shown; fitted parameter estimates with standard errors in parentheses. * Statistically significant beyond the p < .05 level by univariate two tailed test: i.e., the
parameter estimate is greater than twice its standard error.
Only one antecedent variable, teacher time served in their current school
(Tcurt) at the teacher level, was demonstrated as having a statistically significant,
positive influence on teachers’ perceptions of principals’ laissez-faire leadership
(LFt) scores (Research Question 3).
185
The Effect of Antecedent Variables on School Learning Environment and Teacher
Outcomes
The mediated effects model (Hallinger & Heck, 1998; Chapter 2) proposes
that a more realistic understanding of the relation between principal’s leadership
behaviours and school outcomes such as school learning environment and teacher
outcome factors, can be gained by considering the influence of antecedent variables.
Throughout this analysis, the relation between a selection of school level and teacher
level antecedent variables and leadership style behaviours as independent,
explanatory variables has been considered on seven school learning environment and
four teacher outcome dependent variables.
An earlier analysis considered the influence of demographic variables in
accounting for the variance recorded in teachers’ school learning environment (Tables
5.1 to 5.7) and teacher outcome (Tables 5.8 to 5.11) scores. The results of these
analyses are discussed in the proceeding section, where the effects of antecedent
school average level and teacher level demographic variables are considered in
accounting for the variance recorded in teachers’ school leadership, school learning
environment and teacher outcome scores. School average scores for each school were
derived from aggregated individual teacher scores.
School Level Results
School Leadership variables. Six school level antecedent variables were
derived by averaging corresponding teacher level demographic variables across 52
schools. The six school level demographic variables include teacher position
(executive or teaching; Teachs), teacher experience (Texps), school size (student
population; Ssizs), teacher gender (Tgens) and principal gender (Pgens).
School level school size (Ssizs) was indicated as having a negative influence
on teachers’ perceptions of principals’ individualised consideration (ICs) scores. This
indicated that where schools had large student populations, teachers’ perceptions of
principal’s individualised consideration behaviour decreased. At the school average
level, no other demographic variables were demonstrated as having a statistically
186
significant influence on teachers’ perceptions of either visionary (VIs) or laissez-
faire (LFs) leadership behaviours.
School Learning Environment variables. As far as student supportiveness
(SSs) was concerned, teacher experience (Texps) and principal gender (i.e., whether
the principal was a male or a female; Pgens) were both indicated as having a
positive, statistically significant effect on teachers’ scores. In schools that had a
larger proportion of experienced staff compared to inexperienced staff, teachers’
ability to provide student support also increased. Further, teachers’ scores indicate
that they perceive higher student supportiveness scores under the leadership of
female principals. Two explanatory demographic variables indicated a negative,
although statistically significant influence on teachers’ student supportiveness scores.
Teachers’ scores indicated that in schools where there was a larger proportion of
younger staff as compared to older staff (Texps), teacher perception of student
supportiveness (SSs) decreased. Teachers’ scores also indicated that in schools where
there was a larger proportion of male staff as compared to female staff (Tgens),
teacher perception of student supportiveness (SSs) also decreased.
Principal gender (Pgens) was demonstrated to have a negative, statistically
significant influence on teachers’ perception of centralisation (CENs) scores. This
indicates that teachers’ perception was that decision making was more highly
concentrated in the senior executive in schools led by a male principal as opposed to
schools led by a female principal.
School-average level of teacher experience (Texps) was indicated as having a
negative influence on teachers’ perception of innovation (INNs). Teachers’ scores
indicated their perception that adaptability and responsiveness to innovation (INNs)
are lower in schools where there is a larger proportion of experienced staff compared
to inexperienced staff (Texps). Principal gender (Pgens), in respect of female
principals rather than male principals, was also perceived by teachers as having a
positive influence on a school’s innovativeness (INNs).
Two school level, explanatory demographic variables were indicated as
influencing teachers’ perception of achievement orientation (AOs). Teacher position
187
(Teachs) was indicated as having a statistically significant, negative influence on
teachers’ achievement orientation (AOs) scores, while teacher experience (Texps)
was indicated as having a positive influence on these teachers’ scores. These results
suggested that in schools, the push for achievement (AOs) is more likely to come
from experienced staff as opposed to coming from inexperienced staff (Texps).
Teacher Outcome variables. Two school level, explanatory demographic
factors were indicated as having statistical significance for teachers’ perception of
influence (INFLU). Teacher experience (Texps) was perceived to be positively
associated with teachers’ perception of influence (INFLU), indicating a perception
among teachers in schools that more experienced teachers (Teachs) had greater
influence over their workloads (INFLU) than did less experienced teachers.
Teachers’ scores also indicated that teacher perception of influence (INFLU)
decreased in schools where there was a larger proportion of younger staff as
compared to older staff (Tages).
Teacher Level Results
School Leadership variables. An analysis of data indicated that both the
explanatory variables of teacher position (Teacht) and teacher time with their current
principal (TSCPt) were responsible for at least some of the statistically significant
negative variance recorded in teachers’ perception of principal’s vision (VIs) scores.
This suggests that teachers lose their perception of visionary leadership as they are
promoted through the school system (Teacht), and the longer they serve with their
existing principal (TSCPt). Female teachers, however give principals higher
perception of vision (VIs) ratings than do male teachers.
Teacher position (Teacht) and teacher current length of time in school (Tcurt)
were demonstrated to be have a negative but statistically significant effect on
teachers’ individualised consideration (ICs) scores. This suggests that the longer
teachers are in a particular school (Tcurt), and the more senior they are in terms of
their position within their schools (Teacht), the less they perceive their principal’s
individualised consideration type behaviours. This may simply reflect growing
teacher cynicism the longer they stay in a particular school. This is further supported
188
by the positive, statistically significant influence that teachers’ time in their current
school (Tcurt) has an explanatory variable to account for some variance in teachers’
laissez-faire leadership (LFs) scores.
School Learning Environment variables. The analysis undertaken earlier
indicated that student supportiveness (SSs) was negatively influenced by teacher
position (Teacht). This can be interpreted as suggesting that teachers perceive
themselves as being more supportive of students than are administrative staff. This is
perhaps a function of teachers’ regular exposure to students in classrooms.
Two school learning environment response variables, affiliation (AFFs) and
professional interest (PIs) indicated no statistically significant influence from any of
the six teacher level explanatory demographic variables. However, the response
variable centralisation (CENs) was demonstrated to be negatively influenced by two
explanatory variables, teacher position in the school (Teacht) and teacher length of
time in school (Tcurt). Teachers’ scores indicate a perception that executive staff
make less contribution to the decision making process than teaching staff, and may
well reflect a belief among teachers that it is principals who make the ultimate
decisions in schools. Further, these scores indicate a perception among teachers that
longer serving teachers who have remained as non-executive staff have less influence
in the decision making process in their respective schools than do younger serving
teachers.
The school learning environment innovation (INNs) was demonstrated as
being negatively affected by the teacher level demographic explanatory variable of
teacher position (Teacht). This result suggests that teachers perceive that
administrative staff are less able to implement innovation in their schools, and may
well reflect teacher belief that the focus of administrative roles in their schools is in
curriculum and student welfare.
One explanatory teacher level demographic variable was shown to have a
statistically significant, negative effect on teachers’ perception of resource adequacy
(RAs). This result indicates that as staff move from classroom teachers to
administrative staff, teachers perceive that access to teaching resources declines. This
189
finding may be linked to a perception that administrative staff have less time for
proper lesson preparation.
Teacher current time in their school (Tcurt) and teacher gender (Tgent) were
demonstrated as two teacher level explanatory demographic variables that had a
positive, statistically significant influence on teacher perceptions of achievement
orientation (AOs). Female staff and longer serving staff members were perceived by
teachers to positively influence staff in the push for an achievement orientation
(AOs).
Teacher Outcome variables. An analysis of teachers’ outcome scores
indicated the differential effect that teacher level demographic explanatory variables
had on each of the teacher outcome measures. Two explanatory variables, teacher
position on staff (Teacht) and length of time spent in current school (Tcurt) both
demonstrated a negative, statistically significant influence on teachers’ perception of
global satisfaction (SAT) scores. This suggests that the more senior the staff position
in the school (Teacht), and the more time that a staff member spent on the school
(Tcurt), the lower their global satisfaction (SAT).
Teachers’ perception of influence (INFLU) was an outcome measure
designed to examine a teacher’s view of the influence over workloads they held
within their school. Two teacher level demographic explanatory variables were
shown as affecting teacher perceptions of influence at a statistically significant level.
Teacher position within their school (Teacht) indicated a positive effect on teachers’
perceptions of influence, whereas teacher gender (Tgent) was indicated as having a
statistically significant, negative influence on these teachers’ scores. Teachers
perceive that administrative staff have more influence over their workloads than do
teaching staff, and that male staff have more influence over their workloads than do
female staff.
Interestingly, no teacher level demographic explanatory variables were
indicated as having any statistically significant influence over teacher’s perceptions
of effectiveness (EFF). However, teacher perception of control (TCON) was
indicated as being negatively influenced by teacher position (Teacht), suggesting that
190
classroom teachers perceived they had less control over their workloads than did
administrative staff.
Research Question 3: Summary
Research Question 3 postulated that the antecedent background variables of
teacher position on staff (administrative or non-administrative), the amount of time a
teacher has been teaching, the amount of time a teacher has spent in a school, the size
of a school’s student population, teacher’s age, teacher’s gender, principal’s gender
and the amount of time a teacher has spent working with their current principal will
demonstrate differential impact in terms of teacher perceptions of leadership style,
learning environment and teacher outcomes. This is consistent with the mediated
effects model proposed by Hallinger and Heck (1998), which indicated that
antecedent variables will influence school leadership, school learning environment
and teacher outcomes both directly and indirectly.
An analysis of the data revealed the differential influence school- and
teacher- level demographic explanatory variables had on school leadership, school
learning environment and teacher outcome measures. This was demonstrated by the
different parameter estimates obtained for each of the pairs examined, and by the fact
that some response variables indicated an effect with a corresponding explanatory
variable, while some response variables did not.
School Level. At the school level, the relation of antecedent explanatory
variable to response variable seemed consistent and predictable in most cases. As
teacher positions change within schools and teachers are promoted to more
administrative roles (Teachs; for example, to department or faculty head), teachers
perceive reduced influence of achievement orientation (AOs). This may well be
simply a function of the perception of teachers across many schools that time spent
with students in face-to-face teaching is reduced, as administrative positions usually
attract a lesser teaching load. Conversely, as teacher experience (Texps) increases
over time, then teachers’ perception is that student supportiveness (SSs),
achievement orientation (AOs), and perception of influence (INFLU) all increase.
This reflects teachers becoming more at ease and growing in mastery of their subject
191
in their teaching function, and so being able to attend to other outcomes (namely
student supportiveness, achievement and influence) beyond mere survival in the
classroom. An interesting counter to this is the finding that as teachers become more
experienced in their schools, perceptions of innovation decrease. This may simply
reflect an overall teacher malaise in the public education system based on the
expectation of having to do more with fewer resources.
School level teacher data suggested that as school size (Ssizs) increased,
teachers’ perception of their principal’s individualised consideration (ICs)
behaviours decreased. This perhaps reflects the perception that relationships are
harder to maintain in larger secondary schools than in smaller schools. The data also
indicated that as teachers age, (Tages), there is an accompanying decrease in the
perception of student supportiveness (SSs) and influence (INFLU). This perhaps
reflects a certain level of cynicism that many teachers who have worked in the
system for a long time report experiencing after many years. This cynicism is a
worrying trend, given the maturing age of the teaching population in New South
Wales secondary schools (see Appendix 4.3; Table 4.7), and that at the time of the
survey, major industrial action involving teachers salaries was taking place within
many secondary schools. Lastly, at the school average level, teachers perceived that
female staff were less inclined to student supportiveness (SSs) than were male staff.
Teacher Level. At the individual teacher level, many of the results were
consistent with the findings reported at the school level. It was interpreted that
teachers perceived that a change from a teaching to an administrative position
(Teacht) was significantly related to a decrease in the perception of both visionary
leadership (VIs) and individualised consideration (ICs) behaviour. Further, teachers’
perceptions were that staff in executive positions within their school had a decreased
perception of student supportiveness (SSs), centralisation (CENs), innovation
(INNs), resource adequacy (RAs), global satisfaction with leadership (SAT) and
perception of control (CONT). This perhaps is related to loss of idealism that comes
from obtaining a promotion position, the loss of time due to management
responsibilities and the realisation that leadership in a school context involves
juggling competing needs. Conversely, teachers also perceive that promotion
192
positions (teacht) bring about increased ability to influence the daily management of
teachers.
The teacher level data also indicated that teachers perceived that increased
time in their current school (Tcurt) was associated with a decrease in the perception
of individualised consideration (ICs) behaviour and an increase in the perception of
laissez-faire (LFs) style leadership behaviour. An increase in time in a current school
(Tcurt) was also associated with a decrease in global satisfaction with leadership
(SAT). One explanation that could be advanced is that teachers who stay at one
particular school for too long, develop over time a sense of cynicism that comes from
“seeing through” the principal and their leadership style. Teachers also indicated that
an increase in time in their current schools (Tcurt) was related to a perception of a
decrease in centralisation (CENs). The length of tenure and the sense of teacher
staleness that comes from being in a school for a long time period may well affect
this additional factor. Interestingly, teacher time in their current school (Tcurt) was
also positively related to an increased perception of achievement orientation (AOs),
which perhaps occurs as teachers become less concerned about daily survival in the
classroom and more concerned with encouraging higher standards from students.
The data suggested that at teacher level, female teachers (Tgent) were more
perceptive to the visionary behaviours (VIs) of their principals. This same antecedent
variable was associated with an increased perception of achievement orientation
(AOs) and a decreased perception in influence (INFLU) in their particular schools.
Lastly, an analysis of the data indicated a statistically significant, negative
relation between teacher time spent with current principal (TSCPt) and perception of
visionary leadership (VIs) behaviour. Again, this perhaps develops out of the notion
of being in a school for too long a time and working with the same principal over an
extended period of time such that a sense of cynicism develops from “seeing
through” the principal and their leadership style.
However, an analysis at both school and teacher levels of the effects of these
antecedent variables on perceptions of school leadership, school learning
environment and teacher outcomes, indicates that Research Question 3 can be
193
accepted as true, namely that antecedent variables have a differentiated effect on
leadership styles, school learning environment factors and teacher outcomes.
Summary
The purpose of this phase of the study has been to examine the relation
between the principals’ leadership style behaviours and their influence on
perceptions of school learning environment and selected teacher outcomes. This was
done in the context of the mediated-effects model of effective schools (Hallinger &
Heck, 1998; see Chapter 2), from the perspective of the transformational and
transactional leadership paradigm (Bass & Avolio, 1997; Burns, 1978). Several
findings have become apparent as a result of the multilevel model analyses
undertaken in examining these relations.
Firstly, multilevel analysis pointed to the differential effect that each of the
three leadership style behaviours had on each of the school learning environment and
teacher outcome measures. A second, critical finding flowing from this was the lack
of evidence supporting the effect of visionary leadership in the transformational
leadership process. This was both surprising and contrary to the leadership literature
(Bass & Avolio, 1997; Burns, 1978), which suggested that follower internalisation
and adherence to the vision was the highest motivator the leader could employ to
ensure commitment to group goals.
Laissez-faire leadership style behaviour was demonstrated to have more
influence (albeit negative in one case) over the school learning environment and
teacher outcome measures considered in this analysis. As previously stated
(Hypothesis 1.1 and 1.2: Summary), perceptions of student supportiveness (SSs),
affiliation (AFFs) and achievement orientation (AOs) all increased as perceptions of
laissez-faire principals’ leadership (LFs) style increased. This suggested that teachers
banded together in tight groups, supported students through the school welfare
system and pressed for achievement in the void created by a laissez-faire leadership
style. Further, as perceptions of laissez-faire leadership (LFs) style increased,
teachers perceived a decrease in resource adequacy, perhaps reflecting teacher
194
despair coming from the fact that the laissez-faire principal was not providing
teaching resources in line with their students’ needs. The importance of laissez-faire
leadership, given the special loosely coupled tightly coupled (Weick, 1976) nature of
schools as organisations and the teachers within them gives cause to consider other
possible facets of leadership behaviour, including leadership substitutes. The analysis
of the data gathered indicated that, rather than transformational style leadership
behaviours exerting the most influence on contextual factors such as school learning
environment and teacher outcomes, it was laissez-faire leadership behaviours and not
visionary leadership behaviours that influenced outcomes to a greater degree (Kirby,
King & Paradise, 1992).
The importance of the role of individualised consideration as a motivator for
performance was highlighted, especially in the areas of affiliation, global satisfaction
with leadership and teacher perceptions of influence. Analysis indicated that, in at
least these three cases, the individualised consideration leadership behaviours of the
principal played a far more influential role in affecting teacher’s perceptions than did
the visionary leadership behaviours.
A third finding coming from this analysis was that principals could indeed
target their leadership style behaviours to have differential impacts on schooling
learning environment and teacher outcome variables. Further, some leadership
behaviours were demonstrated to be more effective in achieving organisational goals
than other behaviours. This could be very useful in a practical sense if a principal
wanted to target a particular school improvement area, such as student welfare or
staff collegiality. Leadership style behaviours were also demonstrated to have a
differential effect on teachers’ perceptions of the outcomes considered in this study.
Lastly, the antecedent variables were also demonstrated to have a differential
effect, both at a school and a teacher level, on school leadership, school learning
environment and teacher outcome variables. At the school level, the size of a school
in terms of student population has an important bearing on teachers’ ability to
perceive and respond to different leadership styles. The question that has to be
considered, therefore, is what is the optimal size for a school? Teacher experience
also has an influential effect in determining teacher perceptions on a range of school
195
and classroom outcomes. The results indicate that teachers at differing stages of their
careers have differing perspectives on aspects of school learning environment and
other job-related variables. A question that may well be asked is, what is the optimal
time for teachers to stay in the one school? At the teacher level, the results indicated
that teacher position within schools will greatly influence teacher perceptions of
leadership, school learning environment and teacher outcome variables. Likewise,
teachers’ responses also indicated that the length of time served in their school will
also impact, and mostly negatively, on a range of school learning environment and
teacher outcome perceptions. Again, the question of the optimal time of service in
any one particular school has to be considered.
The three leadership styles (vision (VIs), individualised consideration (ICs),
and laissez-faire leadership (LFs)) examined in this analysis, have demonstrated
differential influences on each of the school learning environment variables (student
supportiveness (SSs), affiliation (AFFs), professional interest (PIs), centralisation
(CENs), innovation (INNs), resource adequacy (RAs) and achievement orientation
(AOs)) and the teacher outcome variables (global satisfaction with leadership (SAT),
perception of influences (INFLU), perception of effectiveness (EFF) and perception
of teacher control (TCON). While this is not surprising, the nature and strength of the
relations between variables provide insight into how leadership styles can be
managed in order to achieved desired organisational outcomes, particularly those
related to teacher perceptions and satisfaction.
However, several questions have arisen out of this analysis, particularly
relating to the place of “vision” in schools, how it is formed and transmitted, and the
processes by which principals demonstrate individualised consideration to members
of their staff. The following chapter, Chapter 6, examines these relations and the
issues surrounding them in more detail, from a qualitative perspective.
196
CHAPTER 6
STUDY 2 RESULTS:
A QUALITATIVE PERSPECTIVE ON THE EFFECTS
OF PRINCIPALS’ LEADERSHIP BEHAVIOURS ON
ASPECTS OF SCHOOL LEARNING ENVIRONMENT
AND SELECTED TEACHER OUTCOMES
Introduction
The purpose of Study 2 was to provide a qualitative perspective on that might
illuminate the findings of Study 1, by specifically considering three of the research
questions discussed earlier. Multimethod studies can provide a sense of convergence
between the statistical evidence gathered by survey instruments and the perceptions
and verbal accounts of teachers in situ, that is missing when only one methodology is
employed (Dereshiwsky, 1999). This in turn can allow a greater confidence in the
results obtained from both sources. Multimethod studies also have the potential to
produce divergent findings that can challenge preliminary results. However, these
divergent findings can act to enhance the study’s overall trustworthiness.
Specifically, a qualitative methodology was used to gain additional
understanding into the relation between principals’ leadership behaviours, teacher
job satisfaction and school learning environment, and to investigate those specific
principal leadership behaviours that enhance or erode both teacher job satisfaction
and perceptions of school learning environment. This was done using a semi-
structured interview technique that was designed to “uncover the thoughts,
perceptions and feelings experienced by informants” (Minichiello, Aroni, Timewell
& Alexander, 1995, p. 10).
In terms of the research questions raised in Chapter 3, the following research
questions lend themselves to a qualitative approach (Research Questions 4.1, 4.2 and
4.3):
197
4.1. What do teachers perceive their principal’s leadership orientation to be?
4.2. To what extent do teachers find working in their schools satisfying, and
how does a principal’s leadership style affect teacher perceptions of
satisfaction?
4.3. What leadership behaviours and strategies can a principal employ that
can enhance or erode a teacher’s perception of their school’s learning
environment?
The purpose of this chapter is to report the results of the qualitative phase of
this study. Firstly, procedural issues related to the collection of data are outlined,
including a brief description of the schools and staff involved in the study. Secondly,
the perceptions among staff of the principal’s leadership style behaviours of
visionary leadership and individualised consideration are examined. Finally, issues
related to staff perception of teacher satisfaction and school learning environment are
considered, particularly in light of the influence of the principal’s leadership
behaviour.
Procedures Undertaken
Table 6.1 outlines the conceptual framework around which the qualitative
phase of this study centred. This framework, based on Dereshiwsky (1999), was
initiated by the Research Questions 4.1, 4.2 and 4.3 proposed earlier in the study and
was based on the findings of the quantitative phase and a literature survey (see
Figure 6.1).
Schools were identified that had demonstrated the transformational factor of
“vision” and the transformational/transactional factor of “individualised
consideration”, in order to question teachers and principals about the perceived
influence of these styles on teacher job satisfaction and school learning
environments.
198
Table 6.1
Conceptual framework for the qualitative research
Task Process Action
1. Statement of the problem
Statement of the problem Research Questions 4.1, 4.2, 4.3
2. Research design and methodology
Research questions Open ended, semi-structured conversational interview
3. Population and sample size
Selection criteria for Schools
Selection of low or non-evident Laissez-Faire (LF) cases; Schools in different contexts and stages of development; time available for interview
4. Instrument development Instrument and protocol development
Semi-structured questions based on quantitative findings (Chapter 5) and literature review
5. Data collection Taped interviews, field notes, interviewers’ recollections
School visits to collect data
6. Data analysis Identify emerging patterns and themes
Content analysis for categories of meaning and themes
7. Reporting procedures Tables of evidence; Voices of informants
Findings presented in tables and key quotations
8. Finding and conclusions Propositions Interpretation and discussion of qualitative findings
Sample Selection
School Selection
Fifty-two (52) schools were involved in the initial quantitative phase of this
research project. The results are discussed in Chapter 5. Residual analysis studies
distributed schools along a continuum of low to high ratings of teacher perceptions of
the three leadership factors of laissez-faire (LF), individualised consideration (IC)
and vision (VI). This was useful in identifying potential participant schools to
approach about involvement in this phase.
199
Two criteria were set in the selection of participant schools. Firstly, the main
thrust of the study was to examine the perceived influence of vision (VI) and
individualised consideration (IC) on selected teacher outcomes and school learning
environment. It was appropriate, therefore, to select schools where laissez-faire (LF)
was considered by the staff questioned to be low or non-evident in the principal’s
leadership style.
Secondly, as the purpose of this phase of the study was to describe and
explain the impacts of vision (VI) and individualised consideration (IC) on aspects of
teacher job satisfaction and school learning environment, it was appropriate to
approach schools in different contexts and at different stages of development to
explore these relationships.
A final practical constraint on sample selection was the timeframe available
to complete the interviews. Interviews were to commence in Term 4 of 2000, leaving
only 7 or 8 usable weeks before the Christmas holiday break began. Of those schools
available as defined by the above criteria, several declined an invitation to
participate, because of “busy-ness” related to the specific time of the year. However,
two of the schools contacted were willing to participate in the survey, and a personal
contact in the third participating school helped to secure the opportunity to conduct
this study at that site.
Background of the Selected Schools
School A—was a south coast rural school of more than 1,000 students which
was established in the mid 1960s, and had a stable, long term staff, many of whom
had served in the School for over 20 years. Although the School was not selective, its
academic results over the past decade were impressive, and this had enabled the
school to build an excellent reputation in the local community. The principal, a
female, had led the School for less than 5 years.
School B —was an urban School established only 11 years ago, specifically
to compete with local private schools as one of the Department of Education and
Training’s “Centres of Excellence”. While the School was impressively equipped
200
with resources, it was situated on a small site and housed well over 1,600 students in
an area originally designed for approximately 900 students. The principal, a female,
had led the School for 11 years, since its inception.
School C—was situated in a major northern tableland rural centre served by
three departmental schools and several private schools. The School, with a student
population of more than 1,000, was established in the 1960s, and, along with most of
rural Australia, had been through years of decline. Moreover, the local community
had perceived it as being “on the wrong side of the tracks”, but was now climbing
back with solid academic test results, and a positive profile in the community. The
school had faced major industrial problems immediately prior to the time of the
interviews, where the Teacher’s Union had been agitating for changes to pay and
working conditions. Several respondents mentioned the dispute through their
interviews. The principal, a male, had led the School for 4 years.
Table 6.2 compares each of Schools A, B and C on several demographic
characteristics, including the time the principal had served in the school, the level of
experience of the principal, the type of school and the number of students in each
school at the time of the interviews.
Table 6.2
Comparison of surveyed schools
School A School B School C
Principal—length of time
in school
4 years 11 years 4 years
Principal—gender Female Female Male
Principal—experience as a
principal
First Second First
Type of school Comprehensive Comprehensive Comprehensive
Number of students in
school
<1,100 <1,600 <1,000
School location Rural City Rural
School established Mid 1960s 1989 Early 1960s
201
Informants
Background of Informants
Codes and pseudonyms have been used to describe both the schools and the
informants who participated in this phase of the study.
Table 6.3 indicates demographic characteristics. Five males and five females
were interviewed over the three schools. Between them they accounted for an
average teaching time in the profession of 7.9 years. School A’s informants were the
most experienced, averaging over 11 years experience.
Table 6.3
Demographic characteristics of informants
School A School B School C Total Males 2 2 1 5 Female 2 1 2 5
Gen
der
Total 4 3 3 10
Total Years Teaching Experience
47 18 14 79
Exp
erie
nce
Average Years Experience
11.75 6 4.66 7.9
Administrative Staff 4 2 0 6
Pos
ition
in
Sch
ool
Teacher 0 1 3 4
School B’s and C’s informants had a comparatively low average number of
years of teaching experience (6 and 4.66 years respectively; see Table 6.3). Further,
informants from School C were all classroom teachers, indicating that their
relationship with the principal was not one that would bring them into daily
“executive” contact with the school’s leader.
202
Principals
School A—Alexis (P1) had risen through the administrative ranks to the level
of leading teacher (curriculum) before taking up her first appointment as principal in
School A, four years prior to the time of the interviews. Alexis’ experience had been
in a variety of rural and city schools, and she felt that her main area of expertise was
in curriculum supervision. Alexis believed very firmly that a quality school
experience involved participation in cultural, sporting and community events, and
was willing for her students to try something new. Alexis was reasonably
opinionated, but always ready to consider alternative viewpoints.
School B—Beverly (P2) was a principal of some experience, who had been
appointed to School B while it was being built, and its students housed in another
nearby school. Beverly had a reputation of getting things done, and was appointed to
School B to ensure that this new concept in schooling, a “Centre of Excellence”,
worked. It involved a close partnership between private businesses and the public
education sector. Beverly selected her own senior administrative staff, including
Deputy Principal and Leading Teacher.
School C—Charles (P3) had served as the Principal of School C for four
years. Prior to this he had been appointed for one year as the Acting Principal of the
School while the Department of Education and Training had made a permanent
appointment. Charles had been very pleased to be appointed as the permanent
principal, and had brought new vitality into the school when he transferred to School
C.
Teacher Informants: School A
Respondent 1—Alice (A1) had served in School A for seventeen years,
working mainly in administrative positions within the school’s executive structure.
Her normal position in the school was as the Head Teacher in Visual Arts, but over
the previous twelve months she had acted in the position of relieving Deputy
Principal. This brought her into close daily professional and personal contact with the
Principal. She had direct experience in developing and maintaining a strategic
203
management plan for the school, and had personal involvement in the creation of the
school’s vision and mission statement. Alice was in her mid fifties, and had served
for most of her teaching experience in country schools.
Respondent 2—Alexander (A2) was in his second year of service at School
A, where he held the administrative position of Head Teacher of Mathematics.
Although he was new to the school, he had served as the Head Teacher of
Mathematics in another country school before being appointed to School A. This
appointment, he believed, was a career “step up”, because he had come from a
smaller school further west in New South Wales, with a much smaller staff. As well,
School A had an excellent reputation in the region, and it was well known that many
staff sought appointments to this school. Alexander was in his early forties, and had
only had a limited involvement in strategic planning in the schools he had taught in.
Respondent 3—Allan (A3) was in his mid to late fifties, and had served in
School A as the Head Teacher of the Physical Education department for twelve
years. At the time of interview, Allan had been teaching for thirty-two years, in five
schools, all of them country based. Allan had seen lots of changes in the teaching
profession during his time as a teacher, but now reflected that the nature of the
changes and the pace at which they were expected to be implemented was more than
he could cope with. Allan was looking forward to retiring within 18 months of the
interview.
Respondent 4—Ally (A4), in her mid fifties, had taught in School A since
1984, and had been a teacher for 33 years. Her background was in English and
History, but she had retrained along the way by obtaining a law degree and was now
teaching Legal Studies, Society and Culture, General Studies and the vocational
areas of Retail Sales Operations. Ally’s experience had been in both country and city
schools, and prior to this she had gained employment experience in the public and
private educational sectors.
204
Teacher Informants: School B
Respondent 1—Belinda (B1) was in her mid thirties, and had worked in the
Human Society and Its Environment faculty for six years. This was Belinda’s second
school appointment, as she had transferred to this current school from another local
school. Although she was a classroom teacher only, Belinda had served on several of
the committees of the school, including the prestigious Financial Committee, which
reported directly to the principal on matters of school budgets and discretionary
spending. As well, she had served on the School Council for the previous two years,
where she had had input into the strategic management of the school.
Respondent 2—Barry (B2), in his early fifties, was an experienced teacher
and administrator who had come to the school only two years after it had
commenced classes. Originally he had trained as a teacher, but two years after this he
retrained as a school counsellor, and had spent approximately fifteen years working
in student welfare in various schools. Barry was now in charge of the Student
Welfare program in the school, and as such, was seen by the staff as one of the
school’s executive staff. He only administered a small staff, but its influence was
directly felt throughout the entire school.
Respondent 3—Bruce (B3), in his late thirties, had served in the school for
five years, and had been teaching in the Computing Studies area for just under
sixteen years. Bruce, along with another teacher, shared the position of “Stage Head
Teacher” for students in Years 9 and 10, which meant he was responsible for all the
welfare and extra curricula activities for this stage or level of education in the school.
Although Bruce held a position of responsibility within the school, he wasn’t
considered a part of the school’s executive structure. During the course of the
interview, Bruce indicated that he believed his time at this school was nearly over,
and he was actively searching for a promotion position in another school.
Teacher Informants: School C
Respondent 1—Claire (C1), in her late twenties, was an inexperienced
teacher of 3 years at the time of the interview. Claire had worked in another school in
205
the same town before being transferring to School C. She had served for only three
terms in the school, and had just commenced her fourth term as a teacher of English
and History. Claire had not had any input into the strategic management plan of the
school, and felt that she was still learning the culture and idiosyncrasies of the
school. She had met several times with the principal as part of an induction program
for new teachers, and the principal had visited her in her classroom as part of this
process.
Respondent 2—Colin (C2) was in his mid to late thirties, and worked as a
Human Society and Its Environment teacher in the school. Colin had accumulated
eleven years of experience teaching in country schools, including the school he was
currently serving in. Colin had completed a trade course before entering the teaching
profession in his late twenties. He had held several positions of responsibility over
the seven years he had been in the school, including currently working as the Year 9
Advisor. In this position he worked closely with the executive staff of the school,
including the principal, but he did not consider himself one of the school’s leaders.
Respondent 3—Cathy (C3) was in her mid thirties, and had been at the
school for five years teaching mostly Food Technology in the Technological and
Applied Studies faculty. Cathy needed time off school in the year before the
interview for family reasons, and had spent time discussing her personal situation
with the principal. She felt she had an extremely good relationship with the principal,
and that he had done far more than was necessary to ensure leave she required was
granted. At one point there was a mix up over a medical examination. Cathy related
how the principal sorted out the problem then and there over the telephone,
representing her situation favourably to higher authorities.
The Interview Instrument
The instrument, a series of open-ended questions and prompts, was developed
after reviewing the literature on school leadership and school learning environment,
and analysing the results of the earlier quantitative research.
206
Specific questions were developed to inquire into the following areas of
school leadership behaviours, teacher job satisfaction and school learning
environment:
a. School Vision—including what it is, what commitment respondents feel
towards it; where and how staff see it operationalised; and any perceived
threats to that vision?
b. Individualised Consideration—including questions that explored the
development of a relationship between the respondent and the principal; and
team building behaviours;
c. Job Satisfaction—including the idea of excellence and what makes working
in this school satisfying; and
d. School Learning Environment—including the focus of decision making
within the school; the effect of innovation and change and the acquiring of
resources.
Along with specific questions, a series of prompts was written to assist
respondents in thinking through some of the issues involved. The complete set of
questions and prompts for staff and principals is found in Appendix 4.5. It was
expected that each teacher interview would take approximately 40 minutes, while
principal interviews might take a longer period of time.
Data Collection
Pre Visitation
Telephone contact was made with each of the schools involved in the study,
and arrangements were made for the collection of data. After approval to conduct the
interviews was obtained, a contact person (an executive staff member in all three
schools) was appointed to randomly select and coordinate the schedule of interviews.
Teacher interviews were scheduled during the school day, and included time to be
spent interviewing the Principal.
207
Interviews
On the designated interview day, the researcher arrived at the normal school
starting time, where the contact person had a schedule of meetings arranged. In two
of the three schools, interviews were carried out in a quiet room, away from the
interruptions of school life. In the third school, interviews were carried out in the
staff common room and at teachers’ desks in the staff room. Principals’ interviews
were carried out in the principal’s office in School C, and the interview room used by
other respondents in School A. Unfortunately, the principal at School B was
unavailable at the scheduled time of interview, due to an unforeseen circumstance.
After an introduction and explanation of privacy and confidentiality
provisions (including gaining the respondent’s signed permission to carry out the
interview; Appendix 6.1), the interview commenced with a brief disclosure about the
general direction the interview would take. Permission was sought to tape-record the
interviews, the transcripts of which would constitute the database to be later
analysed.
Interview Protocols
The need for reliability, the mechanics of the interview process (three or four
informants per school), and the time frame over which interviews were to be
conducted, necessitated a strict format for the interviews. This, however, is viewed
by some researchers (Bogdan & Taylor, 1998; Minichiello et al., 1995) as antithetical
to creating an atmosphere where informants feel free to express their thoughts and
express their points of view.
To counter this, a technique of recursive interviewing was adopted, where the
process of “conversational interaction” (Minichiello et al., 1995, p. 80) was used to
encourage informants to see the relationship between one question and the next. To
do this, a series of prompts was developed for each question to encourage further
clarification as the conversation progressed. Appendix 4.5 displays these questions
and prompts, along with their relation to the dependent variables that formed the
constructs under consideration.
208
Moving off into tangents is a problem when using this type of interview
approach. It was addressed by “funnelling” questions and prompts (Minichiello et al.,
1995). This technique starts by having deliberately broader based questions to initiate
the informant’s thinking in the area, and then focusing questions into specific areas.
Data Analysis
The qualitative analysis in this study led to a large amount of data being
generated from the ten teachers and two principals interviewed, and was held in the
form of tape-recorded interviews, interview transcripts, field notes and the
researcher’s own recollections.
The interviews were transcribed into written accounts so that analysis could
continue. It became evident that the data had to be broken up into smaller units for
analysis, and this was done along the broad conceptual lines of the interview
questions. Analysis focused on the principal’s leadership (vision, individualised
consideration), teacher job satisfaction (excellence, teacher perceptions of worth and
value) and school learning environment (important attitudes and values, creating
positive learning environments, decision making and resourcing).
Content analysis of the transcripts revealed that responses could be broken
into themes and conceptual categories within those themes. Further, it was clear that
there were commonalties in teacher responses, indicating similar meanings and
contexts of response. For example, when teachers were asked about the principal’s
approachability, several commented on an “open-door” policy. Given the context of
the response, staff were indicating that the principal was free to discuss matters with
them if the principal’s office door was left in the ajar position. In the minds of the
informants, this was a clear indication of approachability.
Data was analysed through content analysis searching for categories of
meaning which, together, built up themes. The results of this process are presented
using a two dimensional, cross case process (Patton, 1990) with themes on one axis,
and the contextualised data source (either teacher or principal) on the other axis. This
209
allowed the findings from the data analysis to be described and compared between
teachers and principals across the three schools involved in the study, and for
propositions to be developed based on these findings and their interpretation.
Presentation of Analysed Data
The research questions formed the structure around which interviews
occurred. Each research question was supported by a series of interview questions
designed to open up aspects of the research question for deeper consideration. Each
of these research questions represented an area of inquiry around which data was
gathered and analysed.
The large volume of data generated through the interview process
necessitated reducing this information into a more manageable form. A careful
reading of the transcribed interview texts revealed the possibility of identifying
patterns of meaning in the responses. A frequency count of like responses indicated
their importance and relevance as categories of meanings, which in turn built up the
key concepts or themes emerging from the data.
Phrases that were used commonly by the informants assisted in identifying
each theme. These phrases formed themes that contained informants’ responses that
were consistent with a typology. Linking these responses was specific language
usage grounded in relevant contexts that related to a theme. This process of analysis
was complex, as the contextual nature of the responses had to be taken into
consideration, especially where informants seemed to contradict each other.
Consistent phrase usage and patterns of responses by informants were looked for, so
that propositions could therefore be generated. Propositions could be considered
across all themes, to see if more complex associations were present. Once the data
was analysed and reduced in this way, it could be discussed and interpreted. Out of
this process theoretical propositions were developed, reflecting the complex
associations of meaning which emerged.
210
Aspects of Transformational and Transactional Leadership
An analysis of the data revealed a number of key themes within the areas of
school vision, communication of the vision between principal and staff, teacher
satisfaction and the school learning environment.
Data for each of the themes analysed is presented in tables and discussed with
a view to highlighting how informants perceived transformational and transactional
leadership styles, and their implications for teacher job satisfaction and school
learning environment.
School Leadership–Vision
Ownership of and Commitment to the Vision
Teachers and principals were asked a series of questions related to their
perception of the ownership of their school’s vision. The analysis of their responses
is presented in Table 6.4 below.
In Theme 1, the majority of respondents interviewed agreed that their
respective schools have a vision. What was surprising was that the majority of
teachers interviewed did not clearly understand their school’s vision. Either the
vision had not been clearly defined (School C), or teachers had their own
interpretation of that vision (School B), or they were not sure of the details (School
A).
Colin (C2) at School C commented when referring to the vision:
I don’t know whether it’s formalised down in writing. I think we have those mission statements but personally I think they’re just airy-fairy.
Interestingly, the two principals interviewed could not correctly quote their
respective school’s vision statement either, indicating that it was an evolving
statement that had been inherited from a predecessor.
211
212
Alexis (P1), the principal at School A believed:
Well, you’ve got what is called traditional vision—and if your life depended on it, I don’t think most of them (the staff) would be able to quote the actual words. And that’s because I inherited that vision and I don’t agree with all of it.
The result was that although teachers knew about the vision, many
interviewed lacked a clear understanding of the vision in their respective schools.
One long serving staff member, Cathy (C3) had not even seen the vision in the five
years she had been at School C.
One staff member interviewed, Claire (C1), indicated that she was not
committed to the vision of her school. Further, some teachers reflected that
commitment to the vision is, at least in their perception, based on the length of time
they have taught in the school (Theme 2). The more time the staff member has
served at the school, the more allegiance they felt they had to the vision. Bruce (B3)
at School B stated:
…for the people who have been here five years or plus, you’re imbued with it.
Two staff members, Bruce (B3) and Claire (C1), suggested one reason for
their own lack of commitment to their school’s vision was their newness to their
respective schools. This was an interesting comment from Bruce (B3), who had been
in School B for five years and held an administrative position.
Alexis (P1), the principal at School A, pragmatically suggested that while some
staff were committed to the vision, others were not. Belinda (B1) at School B agreed,
perhaps reflecting the reality of low teacher commitment to a vision statement that
was not clearly explained (Theme 1).
213
Responses
Themes Category
Prin
cip
al
(Sch
ools
)
Sch
oo
l A
Tea
cher
s
Sch
ool B
T
each
ers
Sch
ool C
T
each
ers
1. Teachers lack clear understanding of their school’s vision.
1. The vision is not clearly defined and explained 2. I have my own interpretation of the vision 3. I am not sure of the details 4. It is an evolving vision 5. I don’t own the vision, yet 6. I’ve never seen the vision
P1 P3
P1
A3,A4
B2
B1
C1, C3 C2
C1 C2
2. Teacher’s commitment to the vision is a function of the length of time in the school.
1. I’ve been at the school for a long time 2. I am new to the school 3. Some are committed, and some are not
P1
A1, A3 B2 B3 B1
C1
3. Teachers’ ownership of the vision is a function of whether staff were there at the time it was formulated.
1. I inherited the vision 2. The vision was established before some staff were
appointed to the school 3. I had something to do in the formulation of the vision
P1
A1 A1,A3
B3 B2
C1, C3
4. Vision tends not to change in response to changing staff, students and environmental conditions.
1. The vision is not regularly referred to 2. New staff do not get the opportunity to influence the
vision
A1B1, B2
5. School leaders had more chance of having input into vision formulation.
1. It’s the principal’s vision 2. I’m on the school executive and had an input in
formulating the vision 3. I was on a management committee that influenced the
vision formulation
P3
A1
A3, A4
B2
B1
C3
Ownership of and Commitment to the Vision
Table 6.4
21
3
The next theme (Theme 3) suggested that the degree of ownership of a
school’s vision is a function of whether or not the staff were present at the time it
was formulated. Several informants, including Alice (A1), Bruce (B3), Claire (C1)
and Cathy (C3) acknowledged that their school’s vision was established before some
staff were appointed to their schools. Those staff who were present at the time of
their school’s vision formulation (Alice (A1), Allan (A3) and Barry (B2)) indicated
they had something to do with its formulation, and therefore presumably felt a degree
of ownership of it.
Bruce (B3) at School B, when asked of his commitment to the vision framed
it in this context:
I’m probably one of the latecomers so I see it from a different perspective. People who were here initially definitely had the vision, understand it and probably promote it… of the recent group, I think more could be done to promote what the vision of the school is—I think they have a general, hazy idea of what it is… The perception among some informants was that many of their colleagues did
not own their school’s vision because it was formulated before their arrival at their
schools. Both principals accepted the fact that there were differences in commitment
among their teachers, which is not surprising, given that one principal’s commitment
to ownership of the school’s vision was not strong (Alexis (P1) at School A).
Theme 4 continued on from this, suggesting that the vision of two schools
surveyed (Schools A and B) had not changed or adapted to changing staff, students
or environmental conditions. One staff member, Barry (B2) at School B reflected
that the vision of his school had remained unchanged for too long and that it didn’t
reflect the intentions of many of the newer staff arrivals. In the perception of Belinda
(B1) and Barry (B2) at School B, a vision that had not changed in response to
changing staff, student and external environment conditions could not be considered
a viable vision.
Many staff agreed that school leaders had more of a chance of having input
into the formulation of a school vision than teaching staff (Theme 5). It was those
staff in leadership or executive positions within schools who affirmed they had input
into the formulation of a vision. Both Alice (A1) and Allan (A3) were Faculty Head
214
215
teachers at School A, and Barry (B2) was in charge of student welfare and
counselling at School B, and part of that school’s executive structure. Cathy (C3) at
School C believed that the vision was the property of her principal, Charles (P3).
So, while vision has an important symbolic aspect in setting the direction for
a school to travel, many respondents either didn’t know the details of their school’s
vision, or had their own interpretation of it. Teachers who had served for a long time
were perceived to have been “brought up” with the vision, while those newer to their
schools were not. Further, newer staff members didn’t have the chance to have input
into any modification to the vision, which suggested that the vision had remained
unchanged while staff, students and the outside world moved around them. Several
informants viewed their school’s vision as the concern of the principal and the senior
school executive.
Elements of the Vision
The next question to be explored was what the vision actually entailed.
Informants were asked their perceptions of the main elements of the vision.
Responses are summarised in Table 6.5 below.
While some staff interviewed had difficulty in recalling verbatim the wording
of the vision, and some had their own interpretation of what that vision was, the
majority of teachers interviewed and both principals emphasised in Theme 1
excellence and high standards as being key factors that make up the vision.
216
Table 6.5
Elements of the Vision
Responses
Theme
each
ers
Category
Prin
cip
al
(Sch
ools
)
Sch
oo
l A
Tea
cher
s
Sch
ool B
T S
choo
l C
Tea
cher
s
1. Striving for high standards and a high level of excellence
1. We provide a quality experience P1, P3 A1, A4 B2
2. We strive for high academic standards P1 A1, A3 B3 C1 3. We encourage cooperative, independent learning P3 B1, B2
2. Preparing students for the future 1. Preparing students to be able to fit into the workforce C2 2. Preparing students to be able to cope with change A4 B1 3. Preparing students to take their place in society
1. Developing appropriate social behaviours in students C2
2. Preparing students to be able to cope with society C2 4. Concern about broader educational issues
1. Concerns about the daily management of the school C3
2. Concerns about the future of public education C3
21
6
Theme 1 consisted of three categories. Firstly, several informants recognised
that the educational experience provided in informants’ schools was a quality one
(Alice (A1) and Ally (A4) in School A, and Barry (B2) in School B; Alexis (P1) and
Charles (P3)). Secondly, encouragement should be given to attain high academic
standards (Alice (A1) and Allan (A3) in School A, and Bruce (B3) in School B;
Alexis (P1)). Thirdly, informants believed that students should be encouraged to
develop skills in co-operative and independent learning (Belinda (B1) and Barry (B2)
in School B; Charles (P3)). Clearly the emphasis in Schools A and B was on
academic excellence and high standards, at least in the perceptions of the staff
interviewed and the two principals. In School C, Charles’ (P3) school, only Claire
(C1) recognised this as an important element in her school’s vision. Claire (C1) had
previously mentioned her newness to School C, and that she did not own the vision
of her school. Yet, she agreed with the principal, Charles (P3) that excellence and
high standards were an important component.
The second theme (Theme 2) to emerge regarding the elements of vision
included the need to prepare students both for the workforce (Colin (C2) in School
C) and to be able to deal with change (Belinda (B1) in School B).
Colin (C2) in School C saw his school’s vision as broadly encapsulating the
need to prepare his students to take their place in society (Theme 3). Colin (C2) at
School C defined his role in implementing his school’s vision as:
I’m trying to get these kids to adjust socially as well. A lot of these kids lack skills that will allow them to move into the workforce and be active members of society.
The last theme (Theme 4) that emerged involved an interpretation of the
elements of school vision as dealing with issues broader than student outcomes.
Cathy (C3) in School C regarded the future of public education, and concerns about
the daily management of the school as important elements of her school’s vision.
Striving for high standards and excellence, and preparing students for the future
emerged as key elements in the perceptions of many of the teacher informants
interviewed, as well as the two principals.
217
218
Implementation of the Vision
Three broad themes emerged that described how the vision was implemented
in the schools studied (see Table 6.6 below). The respondents’ perceptions were that
vision was implemented through verbalising it, through the practices of daily routine,
and through the modelling behaviours of the principal.
Verbalising the vision throughout the school community is noted as Theme 1
in informants’ perceptions of how the vision is implemented. Staff informants
mentioned several ways this was undertaken, including speaking about the vision
explicitly at staff meetings (Barry (B2) at School B). However, two staff informants
(Ally (A4) in School A, and Barry (B2) in School B) recognised the role their
principals played in verbalising their school’s vision. This was done through
encouraging open lines of communication, and seeking staff, student and other
stakeholder’s opinions. Further, they saw that their principals took the opportunity
regularly to verbalise their school’s vision.
Bruce (B3) at School B reported the many ways the vision was verbalised in
his school:
whether it be at staff meetings on a regular basis, or the Friday meeting…whether it be when people come and go in the school, or the welcome at the beginning of the year, or at the executive meeting, or in fact it’s the daily newsletter or the weekly bulletin that goes out to the students, these are the principal means by which there is a reminder of the ethos and what seems to be important for the school.
219
Responses
Theme
each
ers
Category
Prin
cip
al
(Sch
ools
)
Sch
oo
l A
T Sch
ool B
T
each
ers
Sch
ool C
T
each
ers
1. The vision is implemented by regularly verbalising it
1. The principal encourages open communications in many forms A4 B2 C3
2. The principal seeks the opinion of staff, students and parents C3 3. The principal takes regular opportunities to speak about it B3 C3 4. The vision is spoken about at staff meetings B2 2 The vision is implemented through the daily practices of school life
1. The principal is seen around the school C3
2. Programs and registers are checked regularly A1
3. The vision is constantly checked against reality B3
4. The principal personally invites staff involvement in school activities A1 B2 C1 5. The principal keeps staff accountable in their teaching activities B3 6. The principal encourages us to listen to others P1, P3 7. The vision is apparent through all the daily practices A1 B3 C1, C2,
C3 8. The principal demonstrates trust in the executive structure A3 B3 3. The principal’s modelled behaviour 1. The principal demonstrates that he/she was once a teacher C2 2. The principal is caring and approachable P1, P3 C2
3. The principal deals quickly and personally with threats to the school’s vision P1 A3 4. The principal demonstrated the vision through personal example C2, C3 5. The principal encourages experimentation P3 B3 6. The principal looks for opportunities to proactively promote the vision P3 C3 7. The principal projects confidence A4 C1 8. The principal is always positive C2
21
9
Implementation of the Vision
Table 6.6
However, as revealed in Theme 2, it is in the daily practices and routines of
the school that the vision of the school is implemented. Staff respondents perceived
daily routines and practices included inviting staff participation in school activities,
ensuring that all the routines of the school were consistent with the vision, listening
to staff and developing trust in an executive structure. Interestingly, Cathy (C3) in
School C believed the vision was implemented in the simple practice of seeing the
principal around the school on a daily basis.
Bruce (B3) at School B reported that the vision was implemented through
evaluating the technical components of daily school life. He stated that it came
through looking at teachers’ programs.
Cathy (C3) at School C agreed that the principal emphasised and promoted
the vision through personal supervision of the daily routines of school life. She
stated:
he (the principal) is very concerned that the day to day teaching that goes on is up to date and is done as well as can be expected. He wants everyone to do well, so he checks programs, registers, that sort of thing—he pops into classes from time to time—it’s a way of keeping up to date with what’s happening in the classrooms.
Several informants stated the importance of an executive structure or “chain
of command” as far as implementing the vision. Bruce (B3), a Year Advisor at
School B, when asked about implementing the vision in his school, reflected:
It comes through how we will manage staff matters, certainly from the senior executive and from the executive in general … the vision as presented is in the background but the mission statements are checked, rechecked and evaluated.
Charles (P3) the principal at School C, commented that the “chain of
command” through the executive system provided a mechanism that allowed him to
monitor the school’s progress: He said:
We have a meeting once a week with the executive—I think overall we have a pretty good idea of what the school is on about and the way it is trying to get there.
The modelling behaviours of principals were perceived by several staff
informants as important means whereby the vision was implemented through the
220
schools surveyed (Theme 3). This modelling behaviour was reported in several ways,
including by the principal caring for staff and being approachable (Colin (C2) in
School C), by encouraging experimentation in classrooms (Bruce (B3) in School B),
and by dealing personally and quickly with issues of conflict (Allan (A3) in School
A).
The issue of the principal’s approachability was mentioned by Colin (C2) at
School C as a means used to implement the school’s vision. By modelling
approachability through practices such as an open door policy, the principal
modelled the same care he wanted staff to display to students in the school. Colin
(C2) commented:
I’ve never seen a principal that is as approachable to kids. The kids will go and knock on his door. We’ve had regimes here before where the kids would not go to the front office.
Cathy (C3) at School C believed the principal’s approachability was related
to the demonstrated care shown to community members:
I think he’s a genuine person—sometimes you get the impression that some people aren’t by saying things, or thinking that that’s what you want to hear—I feel he shows real care to students, and parents and staff.
Alexis (P1) was committed to the same degree of approachability. Regarding
her ability to identify with her staff, she stated:
I do have an open door… I do encourage staff to come to my office and I will not turn them away, and give them a fair hearing. I won’t always agree with them. I’ll also listen to what they have to say if they have got a concern, but I won’t necessarily change things. One respondent commented on the importance of being given an open door to
express her views and opinions. Ally (A4) at School A suggested that it was
symbolic of the principals’ approachability, in that she was able to freely present a
viewpoint:
To me, it’s important. I think it matters. It’s always important that you feel you can go to the leader and have a hearing—not necessarily have what you want at the end of the day—but at least be able to put your case forward—to have that contact, so you don’t feel intimidated.
221
Most staff informants didn’t have any difficulty in recognising the varied
forms that implementing their schools’ vision took, and the central role their
principal played in implementing their vision. This role could be as simple as
encouraging open and two-way communications, or as complex as dealing personally
and quickly with any threats to that vision. The two themes that stood out were in the
areas of supervising daily practices and in modelling the types of behaviour. As far
as modelling behaviours was concerned, the staff informants of School C had no
difficulty in recognising this strategy as implemented by Charles (P3), their principal.
The Centrality of the Vision
Staff informants were asked their perceptions of the centrality of the vision in
their respective schools. Staff informant responses are found in Table 6.7 below.
Two contrasting themes emerged, namely, that the vision drives everything that is
done in the school (Theme 1), and that the vision is in the background rather than in
the forefront of their school’s activities (Theme 2).
Some informants believed that their school’s vision was the driving force
behind all that was done in their schools (Theme 1). Allan (A3) of School A, and
Barry (B2) of School B expressed the belief that their respective school’s vision
provided direction for the staff in their daily teaching functions. Furthermore, Alice
(A1) in School A suggested that vision provided a standard by which teachers could
measure their success.
The second theme to emerge was that vision played only a background role in
the functioning of the three surveyed schools (Theme 2). A consistent category
across all three schools was that the influence of the vision was diluted down through
the school’s chain of command structure (Alice (A1) in School A; Barry (B2) in
School B; Claire (C1) and Cathy (C3) in School C). Claire (C1) at School C saw it
this way:
I don’t think it is very central. I think each department has its own influence.
222
223
Charles (P3), the principal at School C commented that the vision’s centrality was to
guide:
I don’t think it’s terribly important because it is there, so I don’t think it is something that we have to work on—our vision. I don’t think we have to say lets work on our vision. Its there and people are aware of what it is and while things are going well and we are going in the same direction I am more than comfortable with it.
A second response in this theme (Theme 2) was that the central importance of
the vision was replaced by the grind of day to day teaching. What was surprising was
that both principals interviewed agreed that the grind of daily school life could
diminish the motivational effect of a vision. Barry (B2) at School B agreed:
Well, it’s important, but I think other factors get in our road now. 1,700 kids, or 1,600 kids, and the daily grind of teaching—it’s not as easy now—we always have classes of 30, and senior classes of 24. And marking and other things would definitely get in the road.
Some staff informants took an idealistic view that school vision was the
central and guiding influence over the teaching and learning activities that took place
in their schools. Other informants viewed vision more pragmatically, questioning the
impact of its influence in terms of the other competing demands, particularly the
influence of individual faculty groupings within each school. Certainly the two
principals interviewed agreed that the influence of the vision of their schools was
diminished as a result of the daily grind of teaching.
224
Responses
Theme Category
Prin
cip
al
(Sch
ools
)
Sch
oo
l A
Tea
cher
s
Sch
ool B
T
each
ers
Sch
ool C
T
each
ers
1. The vision drives everything that is done in the school
1. The vision gives direction to subordinates in the school A3 B3
2. The vision gives a measure how successful the school has been
A1
2. The vision is in the background rather than in the forefront of school activity
1. The influence of the vision is diluted down through the school’s chain of command
A1 B2 C1, C3
2. The central importance of the vision is replaced by the daily grind of teaching
P1, P3 B2
The Centrality of the Vision
Table 6.7
22
4
Research Question 4.1: Summary of Visionary Leadership Results
Research Question 4.1 examined what teachers perceived the leadership style
of their principals to be. The results above suggest that in a school community, the
vision acts to motivate stakeholders by providing its members with some preferred
view of a desired future state. This means that the processes of building and
articulating a vision are important activities for principals to be involved in
(Lashway, 2000). Transformational principals, according to Bass (1985) will be more
successful in articulating a compelling vision for the future and raising the
expectations of both staff and students to the possibilities of the future than are
transactional principals. Transactional principals will be more concerned in the daily
management of their schools as opposed to building and maintaining a vision.
While respondents in all three schools could identify that a school vision
existed, teachers reported that either they weren’t sure of the exact details, or that
they had their own interpretation of it. The use of words like “excellence” and
“quality” in the vision statement was poorly defined and open to interpretation, and
commitment to the vision was seen by some as a function of involvement in its
formation or the length of service in the school. However, most respondents saw that
having a school vision was important in terms of helping to measure the success and
implementation of programs.
Paradoxically, some staff believed that the vision was best implemented
through the daily practices of their schools. Yet, at the same time, some other staff
informants saw that the vision was diluted in the daily grind of teaching.
An important finding to emerge was the central role played by the principal
as the “guardian” of the vision in the perceptions of the informants. It was seen that
the principal transmitted the vision of the school through speaking about the vision
publicly and privately, through encouraging open lines of communications and
through modelling important behaviours that were to be displayed to the rest of the
school community.
225
226
School Leadership—Individualised Consideration
The other significant factor identified by the quantitative study reported
previously was the leadership style behaviour of individualised consideration. This
factor is characterised by leaders displaying individual concern and understanding
for the needs of followers.
A range of questions was asked of informants, designed to encourage
responses that would allow them to comment on the leadership behaviours of
principals that demonstrated important aspects of this factor. Issues such as whether
the leader knows teachers personally, the worth and value they feel as part of the
organisation, and teacher access to the principal, were discussed.
“Do You Feel the Principal Knows You?”
Schools are communities that place an important emphasis on developing
relationships—teachers with students, teachers with teachers and students with
students, and many promote this through their advertised mission statements.
Table 6.8 below summarises informants’ responses to the question “do you
feel the principal knows you?”
All informants interviewed answered in the affirmative to this question. Colin
(C2) at School C went further, when asked whether the principal knew him,
commenting:
Oh yes. Too well. I’d call him a friend of mine.
One staff informant believed her relationship with the principal developed as
a result of working together on school projects. Alice (A1) in School A had a close
working relationship with Alexis (P1) in her role as the acting deputy principal. She
commented:
Oh yes. I am known to her very well, because I have worked with her as the deputy—but also over the years, because I have done a role like the curriculum coordination planning, and those sort of
227
Responses
Themes Category
Prin
cipa
l (S
choo
ls)
Sch
ool A
T
each
ers
Sch
ool B
T
each
ers
Sch
ool C
T
each
ers
1. The principal knows me well 1. The principal knows me very well 2. The principal treats me as a friend
A2, A3 A1
B2 C3C2
2. The principal knows me to a certain extent
1. It took a while for the principal to get to know me 2. The principal knows me in a business type of way
B1 B3
C1
“Do you feel the principal knows you?”
Table 6.8
22
7
228
things, and the timetable. I’ve worked pretty close with her—but I get on well with her.
Some staff informants commented that special circumstances encouraged the
relationship between principal and teacher to develop. One informant, Cathy (C3) in
School C mentioned the special sick leave and extended leave she needed, which
acted as catalysts in fostering this relationship.
Bruce (B3) in School B commented that any principal-teacher relationship
was purely a professional one:
She sees a side of me—yes, but she doesn’t know me personally because it’s a business environment.
An interesting observation is that no informant mentioned that the principal
did not know him or her.
Although a basic level of relationship was maintained as a result of the
commonality of working together in their respective schools, some staff felt that their
principals had got to know them closely through working together on projects, or
through personal interaction related to some special need.
“Do Principals know their staff?”
The same issue was raised when principals were asked, “do you know your
staff?” Table 6.9 summarised principals’ responses.
As far as principal-teacher relationships were concerned, both principals
agreed that knowing staff meant they could keep track of what’s going on around the
school. Again, Alexis (P1) at School A commented:
I have to know what my staff are doing and I have to know my staff.
229
Responses
Themes Category
Prin
cipa
l (S
choo
ls)
Sch
ool A
T
each
ers
Sch
ool B
T
each
ers
Sch
ool C
T
each
ers
1. I know my staff well 1. I was a leading teacher 2. I have to known what they are doing 3. I have an open door policy
P1 P1
P1, P3
2. I do not know my staff well
“Do you know your staff?”
Table 6.9
22
9
230
Strategies employed to know staff
The question of the strategies employed by principals to get to know their
staff was the next issue to be examined. Table 6.10 summarises informants’
responses.
Two themes were important in the perceptions of staff informants. In the first
theme (Theme 1), the personality of the incumbent principal was perceived as having
an important effect that influenced strategies used to get to know staff. Some
informants recognised there was a genuineness about their principal (Colin (C2) and
Cathy (C3) in School C), that their principal showed care (Cathy (C3) in School C),
and they were appropriately sensitive with staff at particular times (Allan (A3) in
School A) as being important strategies. One staff informant, Colin (C2) at School C
was impressed with the honesty the principal had expressed during a recent industrial
dispute. He said:
He’s under pressure here, and he gets in a situation where he’ll say something to the staff that will be inappropriate—he did that during the strikes—people will be ultra sensitive—and he’ll stand up and say—“listen, I’ve made a mistake—I apologise to you.”
The other major theme (Theme 2) to emerge was that the principal had, by
design, employed specific strategies to get to know the staff. Two strategies appeared
as most influential, namely, the strategy of making time for staff and the strategy of
acknowledging effort and work achievement.
The “open door” policy of making time for the staff had an important
relationship building function in all three schools surveyed. Most staff interpreted
this as an indication that the leader placed a high priority on being available and
accessible to teachers as individuals when they required it. The following comments
were found among informants:
I do this everyday—I have no trouble in doing this... Every morning there will be a line up outside (his) door—you just wait your turn. (Colin (C2) in School C);
I would probably just knock on the door and walk in because she has an open door policy—unless she has someone in there, and she
231
Responses
Themes Category
Prin
cipa
l (S
choo
ls)
Sch
ool A
T
each
ers
Sch
ool B
T
each
ers
Sch
ool C
T
each
ers
1. Personality-related strategies 1. The principal was once a teacher, and knows what teachers are about 2. The principal is genuine in dealing with staff 3. The principal is sensitive when dealing with staff 4. The principal shows real care and concern in their interaction with the staff
P1
P3
P3
A1
A3
C3
C2, C3
C3
2. Intentional strategies 1. The principal accepts criticism 2. The principal acknowledges staff effort 3. The principal attends school functions 4. The principal involves the staff in the decision making processes of the school 5. The principal makes time for staff—an “open door policy” 6. The principal networks with staff 7. The principal is supportive of new ideas in the school 8. The principal uses the staff common room as a meeting place with staff 9. The principal works together with staff on school projects
P1 P1, P3 P1, P3
P1, P3
A1 A3
A3
A1
A4
B2 B3
B2, B3
B3
B2 B3 B1
B3
C2
C1,C2, C3 C2
3. Chance strategies 1. It’s luck that the principal uses to get to know staff B1
23
1
Strategies employed by the Principal to get to know the staff
Table 6.10
does have a lot of people—and more likely than not, I would go in and talk to her. (Ally (A4) in School A).
Many staff interviewed regarded the policy of “open door” accessibility as an
important strategy in the principal’s repertoire of creating an environment that makes
them personally known and individually cared for. When comparing past principals
with the current principal, Colin (C2) in School C continued:
The last bloke, I didn’t go anywhere near him... I’ve got no doubt that the last guy used to follow up on your concerns, but, not openly—you weren’t part of the decision making process.
Alexis (P1), the principal of School A, reflected that an “open door” policy
and all it implied meant that other tasks had to be put off, in order to be available
when staff required it. This meant having to find alternative times to undertake the
administrative tasks of the school:
I take work home on the weekends—but it is because of the nature of the job that I can’t get my job done at school, because I’ve got to be available for what ever happens—I put my stuff on hold, so that other things can get done.
A second strategy of acknowledging effort and work achievement was carried
out in many ways, including at assemblies, through newsletters and in the local press,
publicly at a staff morning tea, or privately in the form of a written note.
From the teacher’s perspective, supervisor recognition in public of work
achievement encouraged commitment to strive for more success, while from the
principal’s viewpoint, celebrating the success of one meant that all enjoyed the
benefits. Alice (A1) at School A reflected:
We have a morning tea which is catered for by the hospitality class, and she gives out awards—funny, stupid things like peppermint patty awards—she does that. She does acknowledge staff personally for a job well done, and also we have the staff news each week where she acknowledges staff and their efforts—and sometimes, there are one or two funny, stupid ones in there as well—being a little light-hearted.
Alexis (P1) the principal at School A noted several strategies that she
employed to create and build staff relationships. Open communications, laughing
together as a staff, valuing overlooked staff members, like office and ancillary staff,
and rewarding staff achievement using tokens such as lollies:
232
233
Of course, there’s always mint patties. It’s bit of a tradition, now—and we have a go at each other—someone’s done something pretty silly—at the fortnightly morning tea —we have a “ritual humiliation”. Principals employ a number of strategies to develop relationships with their
staff. Some of those strategies are augmented by the principal’s own personality,
while others are simply procedural directives that staff can use to make contact.
The Principal’s Accessibility
The issue of the leader’s accessibility was raised with staff informants when
asked “is the principal approachable”? Table 6.11 summarises their responses.
The majority of staff interviewed reported that in their perception, their
principal was extremely approachable to talk about both educational and personal
issues. Colin (C2) at School C went so far as to mention:
Well, he demonstrates that by the fact that you can ring him at home at any time and he will talk to you.
Although the majority of staff informants acknowledged that they did enjoy
good access to the principal, several noted that it was “within limits”. If the principal
could not be seen because of scheduling clashes, most informants recognised that
booking in at a later time was a viable alternative.
The first theme to emerge (Theme 1) was the sense of respondents’ ease in
approaching their principals, which related to the principal’s personality. Informants
mentioned characteristics such as compassion (Alice (A1) in School A), respect
(Allan (A3) in School A), trust and fairness (Ally (A4) in School A),
234
Responses
Themes Category
Prin
cipa
l (S
choo
ls)
Sch
ool A
T
each
ers
Sch
ool B
T
each
ers
Sch
ool C
T
each
ers
1. The principal’s personality encourages us to approach them when we need to
1. The principal is business like in approachability 2. The principal is not judgmental when we approach them 3. The principal has demonstrated credibility 4. The principal is compassionate when we approach them 5. The principal is fair
P3
P1
A1 A4
B3
B2
C2
2. The principal has planned strategies that we can follow
1. The principal includes staff in the decision making process 2. The principal is available to staff 3. The principal is both task and relationship oriented 4. The principal is present at school events 5. Staff may not agree with the principal’s decisions, but they do respect the principal for making them 6. The principal talks with students
P3 P1, P3
P1 P1
P3
A1, A3, A4 A1
A1, A3
A3
B2, B3
B3 B2
C2 C1,C2,
C3
C3
23
4
“Do you feel the principal is approachable?”
Table 6.11
and honesty (Cathy (C3) in School C) as desirable personality traits that encouraged
approachability. However, one informant (Bruce (B3) in School B) felt constrained
by the business-like manner of his principal.
The second theme (Theme 2) was that principals had strategies they used to
ensure staff had access to them. These strategies included being present at school
events (Barry (B2) in School B), including staff in the decision making process
(Colin (C2) in School C) and actively seeking student opinion by talking with them
(Cathy (C3) in School C).
Most informants interviewed felt their principals displayed the appropriate
degree of accessibility. As with the strategies employed by principals to get to know
their staff, many accessibility strategies were augmented by the principal’s own
personality. Many staff responded positively to a principal who was fair, honest and
non-judgmental in their interaction with teachers, and felt that these characteristics
could be further enhanced by a principal who was available to speak with them.
Research Question 4.1: Summary of Individualised Consideration Results
Research Question 4.1 examined teachers’ perceptions of the leadership style
behaviours of their principals. A consistent theme to emerge from the data gathered
from informants in the three schools surveyed was that importance was placed on
developing personal relationships as a part of being in an educational community.
The quantitative findings reported in Chapter 5 suggested that principals who
displayed the hybrid transformational/transactional leadership style of individualised
consideration emphasised those behaviours that encouraged the development and
building of relationships. Teacher informants referred to three specific relationships
in reflections on their school communities, including principal to teacher
(leadership), teacher to teacher (collegiality) and teacher to student (pastoral).
The majority of staff informants felt that they were known as individuals by
their principals, and the two principals interviewed acknowledged that knowing their
staff was part of the leadership function they exercised in their respective schools.
Some informants saw accessibility to the principal as their right, while others
235
236
recognised that their principals had applied deliberate strategies to be available when
staff needed them. These strategies included the practice of promoting an open door
policy, which acted as a clear indication that the principal was prepared to spend time
on both personal and professional matters discussing the needs of their staff. From
the principal’s perspective, however, this meant constantly having to put off other
management functions in order to be available when staff needed them.
Findings from the quantitative study reported in Chapter 5 support that staff
are more responsive to the individualised consideration behaviours of principals than
they are to the visionary behaviours of the principal. Further, principals have more
latitude in manipulating school learning environment and teacher outcome variables
using individualised consideration leadership behaviours than they have using
visionary leadership behaviours.
Teacher Job Satisfaction
An important research question posed in this study is examining the
relationship between transformational and transactional leadership styles and
attitudes towards teacher outcomes such as job satisfaction. Bass and Avolio (1994)
argued that transformational leaders could motivate followers to higher levels of
commitment to the vision of their group. However, this study’s quantitative analysis
concluded that higher levels of teacher job satisfaction are associated with the hybrid
transformational/transactional leadership style of individualised consideration.
Strategies used by Principals to Achieve Excellence
Informants were asked about the strategies employed by principals to
encourage greater commitment to achieving excellence in the provision of education
in their schools. Table 6.12 below summarises informants’ responses.
237
Responses
Themes
Sch
ools
)
Category
Prin
cipa
l ( S
choo
l A
Tea
cher
s
Sch
ool B
T
each
ers
Sch
ool C
T
each
ers
1. The principal encourages excellence through personal strategies
1. The staff are acknowledged and valued by the principal, who encourages team work 2. The principal has high expectations 3. The principal leads by example 4. The principal knows what is happening in and around the school
P1, P3
P1, P3 P1, P3 P1, P3
A3, A4
A1, A4 A1, A4
B3
B2
C2, C3
C2, C3 C3
2. The principal encourages excellence through system strategies
1. The principal has developed a culture that celebrates hard work 2. Excellence is the focus of staff professional development and in-service 3. The principal emphasises a chain of command structure of leadership in the school, with clear and open communications 4. The principal encourages a flexible application of school policies 5. The principal emphasises excellence through curriculum and academic supervision
P1
P3
P1
P3
A3
A4
A3, A4
B2
B1
C2, C3
C1
C1, C3
C2
C3
Strategies used by principals to achieve excellence
Table 6.12
23
7
An analysis of informants’ responses suggested two prominent themes,
namely, that principals encouraged excellence through the application of personal
strategies (Theme 1), and that principals encouraged excellence through the
application of system strategies (Theme 2).
With respect to Theme 1, teacher informants believed that their principals
personally encouraged the staff at their respective schools to strive for excellence by
using several strategies, including verbal encouragement (Allan (A3) and Ally (A4)
in School A), taking opportunities to present a positive image about the school in
public forums (Alice (A1) in School A, and Cathy (C3) in School C), and supporting
school events by physically appearing at them (Ally (A4) in School A).
Cathy (C3) in School C referred to the practice that her principal followed in
asking his staff personally to be involved in the extra curricular activities of the
school instead of delegating this task to a deputy principal. Cathy (C3) reflected:
He does come around and ask if anyone can do extra jobs—take on roles, whether it’s on a committee or year adviser or extra catering or sometimes running special things—or special presentations.
Several staff informants noted that their principal led by example, particularly
in the area of building a sense of community in the school. Some staff informants
commented that their principal could be seen both symbolically and practically
leading their schools “from the front”. For Alexis (School A), this meant helping to
sew outfits for student performers in the rock eisteddfod, or, for Beverly (School B),
dressing up in costume to assist in fund raising. Charles (School C) had organised a
working party of parents, teachers and the local community to rejuvenate the front
entry of the school as well as showing up himself to do some of the manual labour
required to complete the task. Knowing that the principal was personally involved in
the project or task added to the weight of staff commitment. Colin (C2) in School C
expressed his delight at seeing the principal coaching a sports team in the school:
The first year he was here—we’ve never seen a principal coach a sporting team before—its just unheard of—and people—some English History people—said how ridiculous is that—but I sat back and said “The kids are going to love this”. . . because he’s willing to get out and do it.
238
Colin also mentioned his principal, Charles’ (P3) egalitarian approach to
leadership, and suggested this encouraged a higher level of personal commitment:
He’s just as willing to go and wash the bus as he is to set the staffing ratios for next year—and that in my view shows that he believes he’s not above us—now that’s important to me—I hate people who feel that they’re running the show and that you’re just the underling.
Some staff informants considered that their principals encouraged excellence
by knowing what was happening in and around their schools. Two informants, Alice
(A1) in School A, and Cathy (C3) in School C believed their principals knew not
only what was happening in their schools, but what was happening in their personal
and professional lives as well. Alexis (P1) saw the utility in knowing her staff well:
I have to know what my staff are doing and I have to know my staff. I may not actually go up to their faculties and have morning tea with them … but I’ll listen to what they have to say if they have got a concern.
Ally (A4) in School A believed that knowing what was happening in her
school was a strength of her principal, Alexis (P1). She reflected:
I think that’s the thing about her. One of her great strengths, from my point of view, is that she is aware of what is done at school by people who don’t necessarily have executive positions, or a very high profile.
Staff informants also suggested that principals encouraged excellence through
developing system strategies (Theme 2). Among these were that principals had
developed a culture that celebrated hard work, and that excellence was a focus of
staff development and in-service programs. Informants also believed that a chain of
command executive structure which emphasised clear and open communications,
flexible application of school policies, and the principal’s supervisory practices, all
encouraged excellence.
Some staff informants (Allan (A3) in School A; Colin (C2) and Cathy (C3) in
School C) believed that their principals encouraged excellence in the way they had
developed school cultures which emphasised and celebrated hard work. Colin (C2)
in School C reflected that his principal, Charles (P3) encouraged excellence by
reinforcing teamwork that emphasised group ownership:
I think the whole idea is that you’re a member of a team at C, and also the fact that they are our kids. If you believe it’s just a matter of
239
leaving at 3.15 (p.m.), then you are not part of the whole school culture. You are not going to excel at your job.
A second system strategy identified by informants as a means used by
principals to encourage excellence was in the area of targeting inservice and
professional development. Charles (P3), the principal at School C, encouraged his
staff to apply for professional development training, even though it was common
knowledge that finances were limited. However, because his focus was on
encouraging students to strive for excellence, he was willing to find other sources of
finance from within the school to support staff in professional development aimed at
realising that focus. He said:
There are a number of professional development activities that they can go on if they want to do professional development for the benefit of their kids. They are certainly not discouraged; it depends sometimes on money to fund but generally we find a way. Staff are aware that they can ask. We get $2,700 a year in our budget for professional development that is $26 for each teacher and that doesn’t go very far…
This focus was particularly relevant to one informant, Claire (C1) at School
C, who compared her current school with the school she came from in the area of in-
service development for staff. She commented:
Well, up until now he has been very good about in-services. I have been to a couple this year and there has never been the case where he has said, no you can’t go, which was the situation where I was last year. So it has been better here.
The development of a strong executive “chain of command” structure of
leadership in the school was seen by some staff respondents as a system strategy
used by principals to encourage excellence. To Barry (B2) in School B, this meant
the principal along with the entire school executive could be relied on to give
teachers support in their daily tasks of teaching. He said:
I think the structure at the top, and it’s not just one of them—it’s the deputies as well—and the rest of the executive—if they can give you that feeling of support and encouragement and willingness to be there when you need assistance, then your credibility and your degree of satisfaction increases.
One staff informant, Claire (C1) at School C, felt that consistency within the
executive structure was an important factor in communicating school culture and
240
ethos, especially in a big school. When asked how the principal encouraged
excellence in her school, Claire (C1) responded:
I think because C is such a big school we tend to deal more with our head teachers than with the Principal. I think through the ethics that are passed through our department.
Colin (C2) in School C commented on another strategy that the principal used
to encourage excellence and greater commitment to the vision of the school. He
suggested that the principal’s flexible application of school policies and procedures
in allowing staff access to school equipment for personal use encouraged
commitment. He reflected:
I think another strategy he uses is . . . we have school buses and school utes and all that type of stuff. People who are pretty good and do their job and that sort of stuff—if you ask him—“I’ve got a bit of rubbish at home—can I just drop it out at the tip”—that sort of mutual type “grease” thing—it might be superficial but it works—people feel “I’m not only just a number here”.
A final system strategy identified by informants as means used by principals
to encourage excellence was in the area of curriculum supervision and targeting
discretionary spending. Cathy (C3) at School C suggested that her principal, Charles
(P3) encouraged excellence by knowing what was happening in classrooms and in
the curriculum. She said about her principal:
He’s always aware of what you are doing through your programs and registers, so he knows on a day to day basis what’s happening in the classrooms.
Cathy (C3) made a further comment that this information could be used to
present a positive image of the school to the public, especially where school
examination results were concerned:
He knows what’s doing and happening in our department. He doesn’t seem to rate some departments more highly than others. He wants to portray what’s happening to the public, so he likes to know what’s happening and that we are doing our jobs as well—I mean the results last year were good—and that’s one aspect of it.
Allan (A3) in School A also believed his principal Alexis (P1) encouraged
excellence in the way she supervised the academic progress of students through the
school, including examining student results from public, standardised tests. He said:
241
We look, for example at the end of the year when our results come in. Alexis will have us in the office just for a short meeting to talk over our results. Were you happy with your results? Do you feel there was a chance of improving on those results? How can we improve on those results? In some areas, it’s obvious that if you had more money, and things come down to the dollar, you had more money, and you’re able to buy more texts, computer rooms, you have all these things, then sure you’re going to get better results. Although principals, in the perceptions of the informants interviewed, used a
variety of methods to encourage their staff to aim for excellence in the course of their
daily teaching, it was the combination of both personal and system strategies which
had the greatest effect in achieving principals’ excellence goals.
How Principals let Staff know that they’ve done a Good Job
Informants referred to “communication” as a systems attribute, which they
interpreted as a strategy that the principal used to tell them they had done a good job.
Table 6.13 below summarises informant responses.
Principals let staff know that they’d done a good job by communicating with
individual staff (Theme 1); communicating within the context of a group (Theme 2);
or communicating by personal example (Theme 3).
A common method of acknowledging good work was simply by the principal
saying so to individual teachers (Theme 1). Principals did this in a number of forums,
both publicly and privately, including through private acknowledgement (Alice (A1)
in School A) or by finding the staff member individually and saying so (Alice (A1)
in School A, Bruce (B3) in School B, and Colin (C2) in School C). Allan (A3) and
Ally (A4) at School A noted that their principal, Alexis (P1) publicly acknowledged
staff contribution in the life of the school. Allan (A3) commented that individual
teachers were often publicly thanked for their contribution:
She always acknowledges people’s effort quite publicly, through our newsletter, (or) in a public forum, such as morning tea. She always acknowledges, and I think that’s the thing about her, one of her great strengths, from my point of view, is that she is aware of what is done at school by people who don’t necessarily have executive positions, or a very high profile. She always acknowledges positive contribution.
242
243
Ally (A4) at the same school (A) agreed, stating:
She’s very, very quick to accept something that is being done well and to congratulate staff—she’s very good like that. If you’ve done something well, she’ll tell you. On the other hand, if you’ve done something unprofessional, you’ll get hauled over the coals.
244
Responses
Themes Category
Prin
cip
al
(Sch
ools
)
Sch
oo
l A
Tea
cher
s
Sch
ool B
T
each
ers
Sch
ool C
T
each
ers
1. The principal lets individual staff know they have done a good job
1. The principal talks to you personally 2. The principal will tell you if something is wrong 3. The principal will communicate by private acknowledgement 4. The principal sends Christmas cards
P1, P3 P1 P1
A1
A1
B3 B2
B2
C2
2. The principal lets staff know as a group that they have done a good job
1. The principal gives formal encouragement 2. The principal uses school gatherings such as an assembly 3. The principal promotes achievement publicly in local newspapers 4. The principal gives staff awards 5. The principal supports staff initiatives 6. The principal makes comments through the faculty head 7. The principal communicates through newsletters to parents
P3 P3
P1 P1
A1
A1
B3 B3
B2
C3
C1, C3
3. The principal communicates through personal example
1. The principal gets involved and models appropriate actions and attitudes
P3
C2
How principals let staff know that they’ve done a good job
Table 6.13
24
4
Other communication forums were mentioned that emphasised the good work
the staff as a whole had undertaken (Theme 2), including school assemblies (Bruce
(B3) in School B, and Cathy (C3) in School C), in the local press (Bruce (B3) in
School B), through weekly staff and parent newsletters (Cathy (C3) in School C) and
by using the executive “chain of command” (Claire (C1) in School C). Alexis (P1) in
School A saw the value of inclusion as far as complimenting staff for work well
done. Of the office and ancillary staff, Alexis (P1) said:
One of the things I particularly like, and I think and know it’s since I’ve come, that the ladies in the front office… are seen as important—they are part of the staff. They don’t feel that they’re there just to do the hackwork and that’s important—they enjoy coming to school because they feel important—they’ve got an importance beyond just a computer.
A third theme was that principals communicated satisfaction with their staff’s
good work through their personal example of getting involved themselves in
activities of their schools (Theme 3). Colin (C2) in School C believed his own
commitment to the vision of the school had increased because he saw the principal
pitching in and doing the work when effort was required. Colin (C3) was impressed
by Charles’ (P3) commitment and action across many areas of school life. This had
many positive effects in the life of the school, not the least of which was on morale
and on raising individual teacher’s perceptions of their worth and value in the school
community. Colin (C2) reflected:
I think he encourages me by making me feel valued here.
Informants saw the act of letting staff know they had done a good job, both
publicly and privately, and individually and in large groups, as a positive strategy
that encouraged their increased commitment to the vision of their schools.
What Staff Believe makes Working in their School Satisfying
The final issue considered by informants was related to what they believe
makes working in their school satisfying. Table 6.14 below summarises their
responses.
245
246
Several themes emerged from an analysis of informants’ responses. These
included intrinsic motivation (Theme 1), collegiality (Theme 2), student relations
(Theme 3), the quality of the school environment (Theme 4), the leadership practices
of the principal (Theme 5) and the role of the parents (Theme 6).
Several staff informants across all three schools involved in this survey
mentioned the role of intrinsic motivation (Theme 1). Three staff suggested that their
satisfaction derived from the fact that they enjoyed the job they did in the school they
were currently working in (Alice (A1) in School A, Belinda (B1) in School B, and
Colin (C2) in School C). Two informants suggested that they believed their job was
worthwhile (Bruce (B3) in School B, and Colin (C2) in School C), while others
mentioned their commitment to the students they currently worked with (Bruce (B3)
in School B, and Claire (C1) in School C). Bruce (B3) in School (B), who had
initiated a number of year group specific programs, reflected:
I believe I’ve done a good job while being here—I’ve initiated a number of brand new things and those things have been beneficial not just for Year 7 and 8 but the whole school. So, from my perspective, that’s been satisfying for me.
Several informants reflected on the quality of collegial relationships that
existed between staff (Theme 2). These included intra and inter faculty relationships,
as well as relationships with ancillary staff. Allan (A3) in School A stated:
The staff are quite easy to work with—there don’t seem to be unpleasant clashes. I’ve got an office as big in the area as this (indicating with his hands) with five people in it. In the 12 years I’ve been there we have never had a cross word. People don’t understand how well we get on together. We work as a team—I like that and I value that. And also I guess, it’s like everything else, even the office ladies are very, very helpful. The staff are very helpful. When you put all those factors together, it makes for a happy environment.
Ally (A4) at School A agreed with Allan’s (A3) assessment of collegial
relations in their school:
I also work with particularly good teachers, and I’m in the staff room with people who I have an excellent relationship with, who feel the same sense of commitment to the kids as I do. So that’s another big issue. I enjoy my immediate workplace, and I suppose it provides me with satisfaction personally. I feel that what I’m doing is valued.
247
Responses
Themes Category
Prin
cip
al
(Sch
ools
)
Sch
oo
l A
Tea
cher
s
Sch
ool B
T
each
ers
Sch
ool C
T
each
ers
1. Intrinsic motivation 1. Teaching in this school has help to improve my teaching 2. Teaching in this school has given me a sense of doing something worthwhile 3. I enjoy the job I do in my school 4. I’m committed to the students in my school
A1
B1 B3
B1 B3
C2
C2 C1
2. Collegiality 1. I work with a great faculty in this school 2. I work with good teachers in this school 3. We work together as a team in this school
P3
P3
A1 A3
B2, B3
B3
C1 3. Student relations 1. I enjoy seeing students develop and achieve
2. This is a good school for kids 3. Students in this school are responsive to any help they are given 4. The students in this school treat me with respect
P1
A1, A4 A3
B1 B2
C2
C2
4. The quality of the school environment
1. The school is a positive and happy place to work 2. In comparison with previous schools, it’s the best 3. This school has a good reputation
P3
P1
A1, A3
B3
C3
5. Leadership practices of the principal
1. The leadership of the school is good 2. The principal of the school is consistent 3. The principal supports us 4. We laugh a lot at school
P3 P1
A1
C3
C2, C3
6. Parents 1. The parents support us in this school B2
What staff believe makes working in their school satisfying
Table 6.14
24
7
The words “team” and “team member” were mentioned a number of times,
and some staff informants reflected that their satisfaction in teaching at their
particular school grew out of an expectation that they were a small part of a bigger
picture. Barry (B2) at School B reflected:
We can rely on other staff to do their job, and they do it well. That makes it a lot easier, and I think it’s valued. I think they have a common purpose…They are certainly working in their departments together (and) we are a big staff.
Relationships with students provided a continuing source of teacher
satisfaction in schools (Theme 3). Many staff informants indicated that their
interaction with the students in their schools was a key reason for them remaining in
the teaching profession. Respondents made many positive reflections about their
interactions with students, particularly in terms of discipline. Alice (A1) in School A
commented:
I think the kids here are really good. Even though we whine about the kids, we don’t have a really big group of kids that do the wrong thing.
Barry (B2) in School B also saw student discipline as providing a positive
satisfier for him. Talking about the students in School B, he stated:
I’ve always liked the students. I was only saying yesterday—we do very little disciplining here…
Other sources of satisfaction mentioned included environmental factors
(Theme 4). Some informants mentioned that their schools were happy and positive
places to work in (Alice (A1) and Allan (A3) in School A, and Bruce (B3) in School
B), and that compared to other schools, their current school was the best.
Two informants in School C (Colin (C2) and Cathy (C3)) recognised the
important influence the principal’s behaviour can have on staff morale (Theme 5).
Colin (C2) compared his current school with his previous school, and indicated that
the principal’s behaviour had an impact on staff morale. He stated:
Well, I’ve recently worked at another school and during that time … the leadership was totally inconsistent—non existent at times—say one thing and do another—not follow it through and I know the staff morale was fairly low…
248
Lastly, one staff informant, Barry (B2) in School B viewed parental support
as a contributing factor that influenced teachers’ perceptions of job satisfaction in his
school.
Research Question 4.2: Summary of Teacher Job Satisfaction Results
Research Question 4.2 examined the extent to which teachers found working
in their schools satisfying, and the degree to which their principal’s leadership style
affected that perception of satisfaction. The data collected from the three schools
involved in this survey suggested that principals’ transformational and transactional
leadership behaviours had a mediating effect on teacher outcomes such as
perceptions of job satisfaction. Staff responses indicated that across several of the
questions raised, teacher perceptions, at least in part, were influenced directly by the
principals’ leadership behaviour.
Whether transformational and/or transactional leadership behaviours of
principals led to differential effects on perceptions of teacher outcomes, needs to be
considered. The evidence gathered in this phase of the study suggests that staff are
more responsive to the personalised and relational behaviours of their principals,
whether that be asking for their opinions on matters relating to school policy, or
being consistent in their dealings with individuals. This finding is reinforced by the
findings of the quantitative component of the study, where a very high correlation
was found between global satisfaction with leadership and principals’ individualised
consideration style behaviour.
The notion of excellence was certainly mentioned in several staff responses to
the question of what their school visions contained (Table 6.5). Of the themes and
categories that emerged from the data related to the area of excellence (Table 6.12),
several emphasised those system strategies that the principal had put in place within
their schools that would encourage greater commitment to achieving the vision. A
key strategy used to achieve this was through close supervision of the curriculum
through monitoring teaching and learning programs.
249
However, it was those relational strategies, such as emphasising team
building, which encouraged staff to feel valued as individuals within their school
communities. Other strategies included inclusively recognising staff effort and
modelling accessibility that staff informants responded to. Two of the themes to
emerge from the question of what made working in their schools satisfying (Table
6.14) related to building a sense of community through establishing and nurturing
relationships—staff with staff (collegiality) and staff with students (relationships
with students). Several staff indicated that they felt known and valued as an
individual because of the individual care demonstrated to them by the principal.
School Learning Environment
A substantive interest in this study is an examination of the perceived
relationships between transformational and transactional leadership behaviours of
school principals and teachers’ perceptions of school learning environment.
Responses were sought from informants to questions about the general school
learning environment, student and staff relationships, change and innovation,
involvement in the decision making process and the adequacy of teaching resources.
Important Attitudes and Values Regarding School Learning Environment
The first issue to be discussed with informants related to those attitudes and
values that were central within the school learning environment, particularly those
held by the principal. Specifically, informants were asked “What attitudes and values
are important in this school as far as the learning environment is concerned”?
Responses are summarised as value statements in Table 6.15 below.
Informants identified four themes relating to values held about their school’s
learning environments. Most responses came from informants in School A, principal
and teachers alike, particularly with respect to their school’s psychosocial
environment. This is not surprising, given the principal’s firmly held beliefs about
what made for a quality school experience (Table 6.15). Valuing teachers in the
250
251
school environment received general support, while values attached to students and
the physical environment were less frequently mentioned in the responses.
A key psychosocial value promoted by Alexis (P1) in School A was that
students could be successful in a variety of different endeavours. Alexis’ attitude was
that all students were encouraged to try a number of different activities. She said:
I want all of the students, boys and girls and the staff as well, to feel if they really want to try something—I want them to feel supported, because that’s how you foster the talent.
Alexis regularly talked up “success” to her students and staff, on assemblies,
in newsletters and at morning teas. High on her personal agenda was encouraging
risk taking in every part of the school community. She defined success in these
terms:
We are such a wonderful school—we nearly always succeed. I really can’t think of anything that I can remember that was not successful—perhaps that’s because I regard success as giving something a go.
Alexander’s (A2 in School A) perception was that many staff had responded
positively to Alexis’ (P1) encouragement to “give something a go”. He reflected:
Staff are willing to be involved in many extracurricular things, and they love to see the school do well—whether it be sport, academic or cultural things—we have had a very successful debating team in the junior school… I think staff value the success that the school has always had and will continue to have…
252
Responses
Themes Category
Prin
cip
al
(Sch
ools
)
Sch
oo
l A
Tea
cher
s
Sch
ool B
T
each
ers
Sch
ool C
T
each
ers
1. Values regarding staff 1. The principal has high expectations of the staff 2. I feel I can rely on other staff—everyone works as a team 3. The staff are very committed to their work 4. The staff enjoy their teaching 5. Learning is supported by resources in the school 6. Staff are approachable 7. Classes should run well 8. All lessons should be meaningful
P3
P1 P1
A1 A1
A3
B3 B1 B2
B3
C1 C2
2. Values regarding students 1. We have a balanced view of students 2. Students are expected to learn in every lesson 3. Students are encouraged to achieve to their potential
P1
A3
B2 B1
C3 3. Values regarding the psychosocial environment
1. We have a positive work ethic 2. Success is acknowledged and celebrated 3. Results engender pride 4. We are successful in a range of activities 5. The school is a happy, safe and secure place 6. We have a positive social environment 7. We have an expectation to work 8. We have a reputation for innovation and success 9. We value learning for learning’s sake
P1
P1 P1
P1
A1 A1,A4
A2
A2,A3
A2,A3
A4
B3
C3
C3
C2
4. Values regarding the physical environment
1. We want to make our school look better 2. Classrooms should be neat and tidy
P3
B2
C1 C1
Important attitudes and values regarding school learning environment
Table 6.15
25
2
Alice (A1 in School A) agreed:
The attitude is that you can succeed, and the idea is that we have always had success, and will therefore be successful. I think that is very important and is promoted that the School has done it again.
Another value that was perceived to be important in the learning environment
of the schools surveyed was that of making the school a safe and secure place, where
students could feel safe from the turbulent, outside world. Charles (P3 in School C)
commented on the socioeconomic circumstances many of his students found
themselves in, and decided to institute a system of visitors’ badges so that unwanted
guests in the school would be easily recognised. This had a positive effect. Colin (C2
at School C) commented:
They know that he (the principal) is concerned and he’s trying to provide the best environment for them in terms of safety and a learning environment—and he stresses that he wants a safe and happy environment for these students. The name badges have really helped.
Allan (A3 at School A) also commented concerning the provision of a safe
and secure environment:
We have tried to ensure there are no outsiders—there are no influences that are going to come in and harm our kids. This is our duty of care. We try to create a safe learning environment for them… I think basically the kids feel comfortable.
With respect to the theme “values regarding staff”, and specifically the
learning and teaching function of staff, principals had high expectations of staff
(Bruce (B3 in School B); Charles (P3) in School C), which in turn led to staff having
high expectations of each other (Belinda (B1) in School B). This was demonstrated
by the fact that a teacher could rely on other staff to teach lessons that were
meaningful and relevant to student needs (Alexis (P1) in School A). Classes were
well run (Alexis (P1) in School A; Allan (A3) in School A) and supported with
appropriate resources and policies that encouraged students to commit to their own
learning (Bruce (B3) in School B; Claire (C1) in School C). Allan (A3) in School A
commented:
When a kid arrives at a lesson, the belief is that there’s going to be a good lesson taught. When they get there, they are expected to work. So that’s one of the first things.
253
Barry (B2) in School B supported this with respect to his school’s learning
environment:
I think it’s very positive—it’s very much a focus. It’s great when you know kids are learning in nearly every place they go to, so it is not a case of you having to draw them back from being in the quadrangle or not in a learning environment.
Values attached to students in the school learning environment focused on
their learning. Categories included the ideals that students were expected to learn in
every lesson (Allan (A3) in School A; Belinda (B1) in School B), and were
encouraged to reach their potential (Alexis (P1) in School A; Cathy (C3) in School
C). Ally (A4) in School B commented:
Well, we try to encourage kids to value learning to start with—learning is important. And also to value themselves as people. And that’s a high priority. We encourage them to do the best of what they do, whatever it might be.
Values perceived as important in the school’s physical environment included
taking care of classrooms, and activities designed to improve the school’s physical
appearance. Claire (C1) in School C summed up the responses by stating:
The classrooms are all in pretty good condition compared to some others that I have seen, and pretty well resourced. They place a lot of value on things like that.
Staff Attitudes to Being in their School
Staff were asked their perceptions of whether or not they wanted to be in their
current school. The specific question asked was “Do the staff in this school want to
be here”? Table 6.16 outlines their responses.
Informants’ major response to this question was “yes”, with their responses
falling into two themes: those unconditionally committed to their present schools and
those expressing conditional commitment. Interestingly, School B respondents
showed most support for the first theme while School C respondents commented
most on the second. The clustering of their responses in this way was consistent with
each school’s profile.
254
255
Informants who indicated that they were unconditionally committed to their
present schools (Theme 1) didn’t wish to speak for other staff in their respective
schools. They felt supported by their teaching colleagues (Alexis (P1) in School A;
Ally (A4) in School A; Belinda (B1) in School B; Claire (C1) and Colin (C2) in
School C), and many of these had been long term appointments in their schools
(Alexis (P1) in School A; Alexander (A2) in School A). Further, they felt supported
in that their school’s chain of command or executive system worked well (Belinda
(B1) in School B), and felt that the only reason why staff left was to seek promotion
positions (Barry (B2) in School B). Alexander (A2) in School A, typified the
perception of staff who wanted to be in their school unconditionally:
Well, the fact that many of them have been here for 15 or 20 years in a large number of cases adds to that—yes, definitely. I think the majority of staff, even though they have their occasional gripes, whines and complaints—they’re human beings—I think in the main they enjoy coming. Concerning staff who were only conditionally happy to be in their current
school (Theme 2), some informants commented that many of their colleagues were
burnt out from the stress of change in the educational sector as a whole (Bruce (B3)
in School B; Claire (C1) in School C). Two informants worried that they were asked
to do more as far as results were concerned with fewer resources (Allan (A3) in
School A; Claire (C1) in School C).
An interesting category that emerged in informant responses across all three
schools was that while teachers felt their schools were good places to work in, they
felt that the education system was not a good place to work in (Alexis (P1), Allan
(A3) in School A; Barry (B3) in School B; Charles (P3), Colin (C2) and Cathy (C3)
in School C). Allan (A3) revealed his attitude to the current state of morale in the
public education sector:
256
Responses
Themes
each
ers
Category
Prin
cip
al
(Sch
ools
)
Sch
oo
l A
T Sch
ool B
T
each
ers
Sch
ool C
T
each
ers
1. Staff unconditionally want to be at their schools
1. Some staff would not want to go elsewhere 2. Staff are supported by their colleagues and principal 3. I can rely on the chain of command in the school 4. Staff only leave when promoted 5. Many staff have been here for a long time 6. My school looks good
P1
P1
A4
A2
B3 B1 B1 B2
C1, C2
C1 2. Staff want to be at their schools, but there are conditions
1. The tone in the school is generally positive 2. Teachers are not held in high esteem in the community 3. Some teachers are burnt out 4. While the school is good, the education system is not 5. There are fewer resources to teach with now 6. Some career changes force moves 7. Teacher self esteem is generally low
P3
P1, P3
A4 A3
A3 A3
B3 B3
C1 C2, C3
C1 C2
C2, C3
Staff attitudes to being in their school
Table 6.16
25
6
257
At the moment, the morale in education is exceptionally low—I have never seen it like this… I start to shake my head about the whole situation. It’s a joke. And the way in which we are perceived, as teachers in the outside community since I’ve been teaching—there are people I don’t even tell now I’m a teacher. There are a lot of teachers who would get out today if they could quite simply because of the trends in education. It’s not (the Principal’s) fault—but, there is no discipline in schools now. We are supposed to be mentors, teachers, parents, and friends. It’s becoming a big ask for a lot of people.
This disillusionment, however, did not carry over into staff feelings towards
their principal. Teacher informants felt committed to their principals (Theme 1,
Category 2) and their schools, and felt supported by the leadership structures in
place in their respective schools (Theme 1, Categories 2 and 3).
Creating and Enhancing a Positive Learning Environment
Informants were next asked about the factors at work in their respective
schools in order to enhance and create a positive learning environment. Table 6.17
summarises informants’ responses.
Several themes emerged in the perceptions of staff regarding the creation of a
positive learning environment. Many perceived that a positive learning environment
had been created and enhanced through celebrations, the classroom, the curriculum,
through students, through principal behaviour, through resources and through the
school’s physical environment.
Celebrating success was seen as an important way of enhancing the learning
environment in School A. Informants recognised that acknowledging achievement
(Alexis (P1), Allan (A3) and Ally (A4)) encouraged participation and involvement,
especially when done publicly, for instance, on an assembly (Alexis (P1), Allan
(A3)).
258
Responses
Themes
each
ers
Category
Prin
cip
al
(Sch
ools
)
Sch
oo
l A
T Sch
ool B
T
each
ers
Sch
ool C
T
each
ers
1. Through celebrations 1. Acknowledging achievement 2. Public displays of celebration, like assemblies
P1 P1
A3,A4 A3
2. Through the classrooms 1. Teachers create stimulating classroom environments 2. Transportable classrooms are no good
A2,A3
B3
C3
3. Through the curriculum 1. Creating an interesting curriculum 2. We negotiate curriculum changes
P1
A1 A1
4. Through students 1. Students are given a voice in running the school through the Student Representative Council 2. Our school has a strong student welfare system 3. We have developed behaviour programs for students
P3
A1
B2,B3
B2,B3
C2
C2 5. Through the principal 1. The principal is seen around the school
2. The principal gives clear communications 3. The principal knows the staff and the students 4. The chain of command leadership structure
P1,P3
P3
B2
C3 C3 C3
6. Through resources 1. Resource allocations are locked in
2. The school obtains good resources 3. Through specific employment—e.g., a gardener or learning difficulties teacher 4. Classes are sometimes held outdoors 5. We don’t have many resources—it’s related to money
P3
A2 A4
A2
B3
B1
C2
7. Through the school’s physical environment
1. Beautifying the physical environment 2. By keeping graffiti levels down 3. By having a building program that advantages students 4. By creating a safe environment
P3
P3
B2,B3
C1,C2 C2 C2 C3
25
8
Creating and enhancing a positive learning environment
Table 6.17
A second theme to emerge was related to providing stimulating and attractive
classrooms. This was done through the decoration of walls with student work,
painting walls in particular colours (Alexander (A2) in School A), and provision of
notice boards and other display equipment throughout the school.
Informants in School A also referred to the theme of creating and enhancing
the learning environment of their schools through the curriculum (Theme 3). This
included negotiating curriculum changes (Alexis (P1), Alice (A1)), and through
creating an interesting curriculum that students would enjoy learning (Alice (A1)).
Including students in genuine decision making processes within the school
was a fourth theme. One particular method employed to do this is through the
Students Representative Council, or SRC. Several informants suggested that this
increased students’ sense of direction and control as partners in the learning
processes of the school. Bruce (B3) in School B suggested that by giving students a
say in the decision making processes of the school, students took ownership of that
decision or project, and saw that it is successfully implemented:
We also give them (the kids) a big say—the kids are actually—the SRC—they have days where, like the big breakfast, they organise the whole breakfast, the end of year activities, the seniors organise and are also involved—there is a lot of ownership and involvement, like the dance—in other words, the kids have a say in the processes of what they want to do. Allied to this was the provision of a strong student welfare network in schools
(Alice (A1) in School A; Barry (B2) and Bruce (B3) in School B).
While it could be inferred that principals’ leadership behaviours contributed
to the school learning environment outcomes in this Table, informants made few
specific references to them (Theme 5) and then mainly from the principal and one
teacher in School C.
The following comment describes how a principal had taken a role in
enhancing the school’s learning environment:
They (students) see him around a bit—not a lot. They are aware of him—they know that he is concerned with them and he’s trying to provide the best environment for them in terms of safety and a learning
259
environment... (Colin (C2) in School C).
A sixth theme identified by informants included the creation and enhancement
of a positive school learning environment by strategically acquiring and using
resources (Theme 6) to enhance the physical environment (Theme 7). Of the small
number of references to this, School C’s anecdotal evidence about the positive
effects of resourcing a particular staff appointment in order to beautify their school
grounds demonstrates how, through the principal’s leadership, an uninviting and
potentially unsafe front entrance to the school was transformed into a showpiece.
The finished effect, including a covered walkway and a large garden was quite
remarkable, and had a unifying effect in the school community. Informants’
comments were very supportive:
Well I mentioned a couple of things before. The walkway has been covered. They have planted trees and things at the edge of the courtyard. Shadecloth areas. The quadrangle. In the classrooms we have just gotten new notice boards.
Whose idea was it to do those things? (Researcher)
The major things I would imagine is the principal. Yeah—certainly it comes from his directive with all the gardens and things. The notice boards I think was a faculty thing but it did make a big difference in the classroom. (Claire (C1) in School C); and
Well, it sounds really trivial, but the fact that we now have a gardener—before, we had grass unmown, we had weeds in the gardens and kids would come in and they would view the place as a dump. (Cathy (C3) in School C).
The positive impact of this work was also felt in the classrooms, at least in the
estimation of staff. Again, Cathy (C3) in School C reflected:
He’s put the shadecloth up there—that’s great, because if the kids are comfortable in the physical environment, and they can see things changing—they can see new gardens being put in—I reckon that that really affects positively in the classroom.
In this case, the Principal was clearly identified by all the staff interviewed in
School C as the agent responsible for creating a positive physical environment for
both staff and students to work in. Charles (P3) was prepared to rearrange School C’s
260
budgetary commitments, and organise both internal and external volunteers to ensure
the work was done. Also, he was prepared to be identified with the project physically
by laying pavers himself during the project.
Another side to beautifying a school’s physical environment was noted in
School B. Bruce (B3) reflected that even the discipline program there was directed
towards making the physical environment of the school better:
Even when we take them on detention, we dig holes and we beautify the environment and, they become owners of that.
School Leadership and the School Learning Environment
The major thrust of this study was to examine the role of principals in
influencing the learning environment of their schools. Three areas were considered
important in examining this influence—the management of change, the decision-
making processes and obtaining teaching and learning resources. Informants were
asked questions relating to these areas, and their responses are displayed in Table
6.18. below.
The effect of the principal’s leadership behaviours on the processes of
innovation and change was considered a source of influence in a school’s learning
environment (Theme 1). The leadership behaviour of the principal was seen as
pivotal in influencing the success or otherwise of innovation and change. It was for
this reason that two informants felt that principals needed to display “strong”
leadership in the management and implementation of change (Alexander (A2) in
School A; Cathy (C3) in School C). Alexander (A2) reflected on the need for
“strong” leadership from the principal when implementing change, given the need
for adaptability of many staff at School A. He stated:
I think that the staff here, given their chronological age aren’t too bad (at adaptation to change)—but it does mean coercion at times. And it means someone in the management side of things has to be able to lead them in that direction.
Cathy (C3) in School C had confidence in her principal, Charles’ (P3) policy
of a “no dramatic changes” attitude which helped staff to be able to cope with any
changes that were implemented. She reflected:
261
262
(School C) is a fairly traditional school—I think Charles is fairly traditional in his views—he doesn’t like dramatic changes, but he does try to stay innovative and up to date and supports us in that.
Two staff informants (Belinda (B1) in School B; Claire (C1) in School C)
believed the executive structure within their respective schools’ chain of command
had an important role in the management of the change process. In School B, Belinda
saw the Deputy Principal as an influential source of change:
I think the Deputy is a good reader —he reads everything—and then he would quietly go to (staff)—we see him a lot because he’s in our Faculty—I think he would go to all the Faculties and try and pass on that information—so that is quite innovative.
Claire (C1) in School C also believed that executive staff had an important
role in managing the change process. It was her belief that any suggestions for
innovation that staff developed would be funnelled through to the executive meeting
and to the principal for consideration. When asked how she would go about
suggesting an innovation, she stated:
I would have to approach my Head Teacher and discuss it with them and approach other teachers who may be interested. It would be presented as a suggestion at the executive meeting, which they have once a week.
Following on from the theme of school leadership’s role in change
management (Theme 1), the role of school leadership in the decision making process
(Theme 2) was considered. The first category was that the principal had the power of
veto over any decision-making processes carried on in their respective schools.
263
Responses
Themes Category
Prin
cip
al
(Sch
ools
)
Sch
oo
l A
Tea
cher
s
Sch
ool B
T
each
ers
Sch
ool C
T
each
ers
1. Leadership in change management 1. Change needs strong leadership 2. Change sometimes means coercion 3. The principal espouses traditional views such as no dramatic changes 4. The chain of command structure encourages implementation of change
P1
A2 A2
B1
C3
C3
C1
2. Leadership in the decision making process
1. The principal has the final power of veto in decision making processes in the school 2. Staff can participate in decision making by holding a position on the executive committee 3. The chain of command structure in the school is extremely important in decision making
P1
P1
A3
A2
A1, A2,
A3, A4
B1, B2, B3
C1, C2, C3
3. Leadership in obtaining teaching and learning resources
1. The principal’s decision to purchase resources is influenced by the school executive 2. The principal makes the decision
P1
P1
A4
B1, B2
C1,C2,C3 C2
School leadership and the school learning environment
Table 6.18
26
3
Alexis (P1) in School A justified her position as the final authority in the
decision making process on several levels. She had been appointed as the Principal for
precisely this task, and the final responsibility for decision making in the school rested
with her position in the school. To Alexis (P1), authentic decision making required a
high degree of accountability. Although many of her staff wanted to make decisions, in
her mind, at least, not many wanted to accept the responsibility for those decisions that
were reached.
However, she explained that she actively sought the opinions of her executive
and other staff members, and, depending on the nature of the decision being reached,
required a fair proportion of the staff to be in agreement with the decision before an
innovation was implemented. On one hand Alexis (P1) displayed a participative
approach to decision making among her staff, while on the other hand she was very
autocratic where she, as the principal, would be held accountable for those decisions.
Two other categories were significant in light of informants’ earlier responses to
change management questions. These were the categories of the role of an executive
committee structure (Category 3) and staff involvement in decision-making (Category
2).
All informants interviewed identified that a clear “chain of command” structure
(Theme 1) assisted them in having input in the decision-making processes of their
schools (Charles (P3) in School C; Alice (A1), Alexander (A2), Allan (A3), Ally (A4) in
School A; Belinda (B1), Barry (B2), Bruce (B3) in School B; Claire (C1), Colin (C2),
Cathy (C3) in School C). Heads of Departments, Deputy Principals and Leading
Teachers were perceived to be conduits that funnelled opinion back to an executive
level, where the decision making could be undertaken. Even though most informants
recognised that the principal had the final right of veto over all decision making
(Category 1), informants were satisfied that their voices would be heard through this
executive structure.
264
Belinda (B2) in School B also acknowledged the right of authority to make
decisions that rested with the principal. In her perception, the chain of command
structure acted to inform the principal in the process of making decisions. When she was
asked, “where are the overall decisions made for the school?” she replied:
With the principal. But she (the principal) would certainly take advice from the deputies, and maybe even the KLA head teachers.
Other informants commented on some of the information gathering techniques
used in their schools. Again, the chain of command funnel was referred to (Category 3),
and the information gathered was passed on in a more formal, “survey” manner. Claire
(C1) in School C reflected:
I know my Head English Teacher is very good with things like that and she has to fill in a survey… she will discuss that with us at a faculty meeting to get our opinions before she hands it on.
As well, most informants interviewed believed that their opinion counted, and
that it was taken into consideration in the decision-making processes of their respective
schools. Specialised committees had been established in several schools to assist in the
process, and other avenues, such as general staff meetings were available if staff wanted
to be involved (Categories 2 and 3). Alexander (A2) in School A commented on the
levels available for staff participation in decision-making in his school:
Well, there are multi levels. We have executive meetings once a week where quite a few decisions are made, which we feel are appropriate to that level of discussion. We then have staff meetings where we feel the staff should be given the opportunity to have input. So, there is that avenue. Then there is a more subjective level, or curriculum level (where) there are the faculty meetings. Every member of staff has opportunity to have input at a staff meeting. And then, the head teacher can take it upon himself to take any further discussion to executive.
The faculty structure is a major avenue for their participation, but other areas
exist, such as the committee structure. Further, general staff meetings are also available,
but many informants suggested that this was not an ideal place for decisions to be made.
Alice (A1) in School A was one of these:
265
We have weekly staff meetings—but I don’t think staff meetings are good places to make decisions. There are too many people, and many people don’t want to be there.
A final theme in this section related to the role of school leadership in obtaining
teaching and learning resources (Theme 3), which was suggested by several informants
as a source of influence on their respective schools’ learning environment. Informants
believed that how principals allocate funds for teaching and learning resources was
influential in determining how the school’s learning environment could best meet the
educational needs of students. Belinda’s (B1, in School B) perception was that the
principal acted as the conduit related to the obtaining and distribution of resources in her
school. This is hardly unexpected, given the nature of School B as a “Centre of
Excellence” and it’s having special school/local industry links. Belinda (B1)
commented:
The principal is definitely the holder of the purse strings—if it was money, and I thought she had it, I would go straight to her and she would give you advice on where to go, and she is also very good at pointing you in the right direction if maybe she cannot provide the funds. In the past we have asked for shelters—so we went to the P and C, but that was with the principal’s blessing—I think you go to her (the principal), and she will point you where to go.
Some informants saw that even though the P and C could provide extra funding,
it was still the principal who directed the flow of finances within their school (Theme 3).
Research Question 4.3: Summary of School Learning Environment Results
Research Question 4.3 examined the leadership behaviours and strategies a
principal could employ that could enhance or erode teachers’ perceptions of their
school’s learning environment. Kirby, King and Paradise (1992) have argued that
transformational school leaders are able to manipulate their environments to meet their
desired goals and outcomes. Theoretically then, the transformational and transactional
leadership styles of principals will have a differential effect on the learning
environments in their respective schools.
266
The informants surveyed in this study could identify those areas of their school’s
learning environment that had been specifically targeted by the principal in the exercise
of leadership responsibilities in order to meet desired outcomes. These areas included
both the promotion of specific attitudes and values in order to create a desired
psychosocial environment, and strategies designed to foster clear lines of
communications and a supportive executive structure.
However, three areas where staff readily identified the influence of the principal
were in change management, decision making and the acquisition of teaching and
learning resources. In these areas staff identified the transformational behaviours of
principals, for example, lifting commitment of staff in programs of innovation by clearly
communicating that achieving organisational goals was impossible without staff
participation. At the same time the hybrid transformational/transactional leadership style
of principals was also recognised, for example, by the fact that the principal knew
individual staff and students and could ask their opinion in matters relating to the daily
management of the school.
Qualitative Findings on Transformational and Transactional
Leadership Styles
The centrality of the vision in the study of the transformational leader has been
well documented in both the corporate world (Bass, 1985a; Bennis & Nanus, 1985;
Burns, 1978; Conger & Kanungo, 1988) and in the educational sector (Kirby, King &
Paradise, 1992; Leithwood & Jantzi, 1997; Leithwood & Jantzi, 2000; Sagor, 1991). So
central is the process of operationalising the vision to the transformational principal, it
becomes the critical leadership function because through the vision, the community is
offered a desirable and preferred future state. Chui, Sharpe and McCormick (1996, p.
32), commenting on three major studies of transformational leaders agreed, and noted
that “vision is the vital attribute of effective transformational leaders”.
It is surprising, then, that the transformational leadership behaviour of vision
formation was not a higher priority among the two (of three) principals interviewed as
267
part of this study (Table 6.4). Both principals (Alexis (P1) and Charles (P3)) indicated
that their current school’s vision was either an inherited statement, or didn’t really
contain the sentiments or phrases that they would support. These two principals further
explained that it was their belief that many of their staff would not be able to correctly
quote their respective school’s vision. Data gathered from staff informants indicated that
this belief was correct, and several staff suggested that they had their own interpretation
of their school’s vision anyway (Table 6.4).
Several other factors were suggested as possible reasons why individual school
visions hadn’t become owned or internalised by staff. These included lack of input at the
vision formulation level, that vision hadn’t changed or adapted to changing
circumstances and staff changes, or that the vision hadn’t been clearly explained and
defined to staff in the first place (Table 6.4).
Even though ownership of their school’s vision was low, many informants still
believed that they knew the domain that their visions included. Words like “quality”,
“success” and “excellence” were used freely by staff informants to describe the elements
of their school’s vision (Table 6.5). However, these words were not clearly defined by
school principals, resulting in broader definitions being applied to these words than
perhaps was originally intended.
Clearly, if vision formation and building are critical functions of transformational
leaders, then principals seeking to raise the commitment of their staff to levels beyond
what would normally be expected, need to attend to those issues which would challenge
the validity of those visions. This would include providing opportunity for stakeholder
input at the vision formation level, seeking avenues to be more inclusive as far as new
stakeholders are concerned, and being flexible with the vision so that it is seen as
adaptive when circumstances warrant it.
So, how did staff become aware of vision within their schools? Even though staff
reported that they were aware of what their school’s vision contained, both principals
and staff recognised the diluting effect the daily grind of school life could have on its
268
implementation (Table 6.7). To counteract this, both principals interviewed suggested
they kept vigilant supervision over the technical aspects of teaching, including
supervision of teaching programs and registers, as well as keeping faculties accountable
as far as public examination results were concerned. Further, the principals interviewed
suggested that maintaining open lines of communication through the executive structure
also assisted in implementing vision.
Another finding to emerge related to the guardianship role that transformational
leaders exercised in terms of protecting their vision from the influence of distraction.
Lashway (2000) argued that in the busyness of daily school life there is rarely enough
time to reflect on what a vision could be, let alone engage in productive dialogue about
alternative visions. Further, Vandenberghe and Staessens (1991) reported that teachers
in schools with poorly defined visions will often replace the school vision with a
personal vision. This certainly was the case for many informants, who reported not
understanding the vision, and therefore replacing it with one of their own (Table 6.4).
One of the key strategies reported by informants when asked how the vision was
transmitted in their respective schools was through the principal’s leadership behaviour
of modelling (Table 6.6). In many respects, this modelling behaviour demonstrated by
the principal provided a link between an unclear or miscommunicated vision statement
and what principals meant when they expressed that their vision was for “a quality
educational experience” in their schools. This cascading effect, from principal to staff,
and then from staff to students provided a clear demonstrated model of the type of
community the principal in each of the schools desired. The principal demonstrated
vision by being highly approachable, dealing quickly with issues as they arose,
encouraging experimentation, maintaining an open door policy, facilitating clear lines of
two-way communication and seeking advice in the decision making process from a
number of the stakeholders within the community.
In the three schools studied, the principals’ modelled behaviour became a close
substitute for an unclear or out of date vision. Leithwood, Jantzi and Steinbach (1999, p.
269
80) described the process of modelling as a powerful tool that could be used to promote
individualised consideration, where the leader “acts as a role model, leading by doing
rather than only telling”. These transformational practices modelled by the principal set
clear examples that staff could follow that were consistent with values and attitudes the
leader espoused (Kirby, King and Paradise, 1992). Principals applying this modelling
strategy welcomed critical feedback, responded positively to teacher initiated innovation
and showed commitment for stakeholder involvement in the decision making process.
The alternative leadership style under examination in this study is the hybrid
transformational/transactional behaviour of individualised consideration. Bass (1985)
originally conceptualised these transformational and transactional behaviours as lying
along a continuum, with the transformational behaviours being the result of enhancing
and elevating the effects of the transactional style behaviours.
At the mid point of the continuum lies the transformational behaviour of
individualised consideration, and the transactional behaviour of contingent reward.
Leithwood, Jantzi and Steinbach (1999) have suggested that both the contingent reward
and individualised consideration behaviours can mesh together, especially in public
educational settings, where the use of monetary rewards is limited. In Leithwood et al.’s
study (1999, p. 73) this combined factor was defined as occurring when the leader
provided “informative feedback about performance in order to enhance a teacher’s sense
of professional self-efficacy”. In this case, opportunity for individualised attention from
the principal was aimed at improving performance and acted as a motivator that raised
teacher expectation about reaching goals, and in turn contributed to a teacher’s overall
sense of satisfaction. In this sense, individualised consideration can act as a substitute for
vision because of its potentially transforming effects.
The quantitative results identified in Chapter 5 support this view of the
effectiveness of individualised consideration. Many of the informants interviewed in this
study had no difficulty in identifying several principal behaviours they perceived as
promoting individualised consideration. A pervasive behaviour in the perception of
270
many informants was the principal’s ability to show a strong sense of caring for the
students and staff who were part of the school community. This care was demonstrated
with the principal’s active support for teachers’ work, especially in the area of teaching
and learning (Table 6.10). Informants suggested that this care was also shown in the
principal’s genuineness and sensitivity when interacting with staff and students.
Leithwood, Jantzi and Steinbach (1999, p. 73) found that principals who “develop a
close knowledge of their individual colleagues” could often use these developed
relationships as the starting point for implementing school wide reform.
Other individualised consideration strategies applied by the principals surveyed
included being seen around their schools, taking interest in the personal and professional
issues of their staff, and pitching to get tasks done (Table 6.11). The strategies of being
accessible, promoting and encouraging open lines of communication, an open door
policy and recognising the efforts of the members of the school community were also
mentioned as important in the perception of staff. These strategies encouraged inclusion
and cohesiveness in respective school communities, and specific individualised
consideration strategies such as including staff in the decision making process in the
school acted to increase staff sense of worth and value.
These transformational and transformational/transactional strategies were
reflected in informants’ perceptions of their job satisfaction (Research Question 4.2).
Informants reflected that among the many factors that influenced their perception of job
satisfaction, the principal’s leadership behaviour of being consistent and supporting
teachers in the task of teaching was important (Table 6.14). Results reported in the
quantitative survey (Chapter 5) indicated that it was the individualised consideration
style behaviours of principals at the teacher level rather than the vision style behaviours
that were statistically significant determinants of teacher perception of overall
satisfaction with leadership. Informants reported that the individualised consideration
behaviours of leading by example, acknowledging and valuing staff, and creating a
culture that celebrates hard work (Table 6.12) encouraged staff to a higher commitment
to excellence.
271
Frieberg and Stein (1999, p. 11) defined school learning environment as “that
quality of a school that helps each individual feel personal worth, dignity and
importance, while simultaneously helping create a sense of belonging to something
beyond ourselves”. Here, Frieberg and Stein (1999) have identified that both visionary
and individualised consideration leadership behaviours can have an influence in
determining the learning environment of a school. Further, no one single factor can
determine the learning environment of a school. Instead it is made up of a series of
school level and classroom level interactions that “create a fabric of support that enables
all members of the community to teach and learn at optimum levels” (Freiberg, 1998, p.
23).
Staff informants indicated that their respective principal’s leadership style
impacted differentially on aspects of their school’s learning environment. Some of the
strategies that emerged from an analysis of the qualitative data came out of a sense of
the principals’ vision for their school, and could therefore be classified as
transformational behaviours. Still other strategies were clearly the result of the
principals’ desire to promote a sense of community and emphasised individualised
consideration behaviours.
Among the transformational behaviours that positively influenced school
learning environment were those that emphasised increased professionalism among staff
as far as the core teaching and learning functions of the school were concerned. Data
analysed in the quantitative survey (Chapter 5) also supported this contention. Pasi
(2001) argued that the role of an effective school leader in promoting a positive learning
environment should include the strategy of encouraging innovation. This, he argued,
created new and interesting opportunities to make learning more stimulating, and
encouraged more students to become engaged in the learning process. In the schools
surveyed, principals held high expectations of the staff, supported teaching and learning
with resources (Table 6.15; Table 6.18) and found ways to direct limited finances to
272
projects deemed important to the teaching and learning functions of the school, and
particularly to supporting the professional development of staff.
However, as discussed earlier, data analysed from the schools surveyed indicated
that individualised consideration behaviours had become close substitutes for a
miscommunicated or unclearly stated school vision. These individualised consideration
behaviours, although a combination of transformational and transactional behaviours
(Bass, 1985a) still had the propensity to act as motivators for many staff.
Several individualised consideration behaviours were mentioned across the three
schools surveyed, and those that were most influential in creating a positive school
learning environment targeted the community building aspects of school life. Leadership
behaviours such as celebrating success (Table 6.15) as a community, both publicly and
privately, helped build a strong sense of community, as did efforts at creating a safe and
secure environment (Table 6.15). Principal behaviours aimed at enhancing the quality of
relationships were also mentioned by informants, including the strategies of using an
open door policy and maintaining open lines of communication (Table 6.17).
In change management and the decision making process, staff informants in two
schools noted the importance of a chain of command structure as a conduit for
supporting and administering innovation. Earlier, staff informants had reported that the
chain of command (or executive structure) in their schools was useful as a
communications funnel that encouraged participative involvement in running the school.
Many informants reported that they felt they had real input into the management of their
schools because of this executive structure (Table 6.18), indicating the individualised
consideration effect of such as structure.
The results of the interaction of these transformational and
transformation/transactional leadership behaviours were that most informants were
happy to be teaching in their current schools with their current principals (Table 6.16).
Those who were only conditionally happy in their present schools with their current
273
principals pointed to demotivators which were outside the sphere of influence of their
principals, such as changing community attitudes towards teachers and a general malaise
with the state educational system generally.
Summary
This chapter has presented the findings obtained from an analysis of principal
and teacher interviews carried out in the qualitative component of this study. The
concluding chapters synthesise these findings and the results obtained from the
quantitative study in Chapter 5 to present the main findings of this study.
274
CHAPTER 7
DISCUSSION AND IMPLICATIONS FOR FURTHER
RESEARCH AND EDUCATIONAL LEADERSHIP
PRACTICE
Introduction
The central objectives of this current study were to utilise a multimethod
approach to: a) extend recent advances in leadership theory to educational
organisations within Australia; b) identify the differential effects of principals’
leadership style behaviours on school learning environment and selected teacher
outcome constructs; c) examine differences between classroom and school level
effects of principals’ leadership style behaviours on school learning environment; d)
identify those strategies that have the potential to enhance and erode teachers’
perceptions of school learning environment and selected teacher satisfaction
outcomes; and e) identify practical strategies that enhance both school learning
environment and perceptions of teacher satisfaction.
The purpose of this chapter is to discuss and synthesise the key findings for
Study 1 and Study 2 within the framework of the study aims and the results of
previous research. Firstly the findings are discussed in the context of each of the aims
of the study. Secondly, the results of Study 1 and Study 2 are synthesised to develop
a more comprehensive and detailed picture of the leadership processes occurring in
the schools under investigation. Thirdly the strengths and limitations of the current
research are identified. Finally the implications of the findings for future research,
theory and educational leadership practice are summarised.
274
Discussion of Findings
Discussion of Aim 1: Testing Differential Leadership Style
Aim 1 of the study was developed to test whether Bass et al.’s (1997)
differential leadership style behaviours that underpin their transformational
leadership model can be identified in a sample of New South Wales (NSW)
secondary schools. This study has applied Bass’ (1985; Bass & Avolio, 1997)
transformational and transactional leadership paradigm to an examination of the
impact of the leadership style behaviours of secondary school principals on school
learning environment and selected teacher outcomes. The starting point for this
investigation was to determine if Bass and Avolio’s (1997) nine-factor leadership
model could be supported from data collected in secondary schools across New
South Wales.
The results emanating from exploratory and confirmatory factor analysis
failed to find support for an underlying nine-factor structure. However, support was
found for a transformational leadership factor (vision), a hybrid low
transformational/high transactional leadership factor (individualised consideration),
and a non-leadership factor (laissez-faire leadership). These results indicate that
respondents in selected New South Wales secondary schools could not differentiate
between the nine factors conceptualised in Bass et al.’s (1997) nine-factor leadership
model. However, the results do suggest that at least three types of differential
leadership behaviours are present, and they imply that the nature of leadership styles
represented in selected New South Wales schools may be different from those found
overseas. New South Wales respondents were only able to identify extreme
transformational and transactional leadership characteristics, along with laissez-faire
leadership styles. The results also offer important insights into the nature of these
three types of leadership behaviours, as is discussed further in the following sections.
Vision. One factor identified through confirmatory factor analysis as a
leadership factor within selected New South Wales schools was vision. In addition,
the qualitative phase of the study illuminated aspects of vision building within a
275
school context, and the extent to which teachers had a clear understanding of the role
of vision within their respective schools. It was particularly evident from the findings
of Study 2 that a school’s “vision” is more than just words in a written statement
(Lashway, 2000). All three schools in the qualitative survey had a vision statement,
as required by the New South Wales Department of Education and Training. Public
school principals are required to be “developing the school’s vision, priorities and
targets which are reflected in the annual school plan” and then “translating the
school’s vision, priorities and targets into explicit policies and practices” (New South
Wales Department of Education and Training, 2000). While a formal document that
represented a vision statement had been created, it was clear among the schools
surveyed that the vision was either an inherited statement, or that it did not have the
input of the current principal or the support of the majority of the current staff.
Tellingly, that vision was rarely referred to.
Several educational writers in the area of vision have emphasised that vision
statements need to be evaluated and updated every three to five years (Lashway,
2000; Sashkin, 1988). Many staff in the present investigation reported that the vision
of their schools had been established before their arrival. Further, many were not
aware of what their school’s vision was, and some of those who were aware of the
vision were not committed to it. These results suggest that while the words of a
vision may have been in place, commitment, understanding and the inclusiveness
that comes from participating in its development were missing. Respondents across
the three schools surveyed also suggested some ways that principals could transmit
their vision imperatives. These included modelling behaviours such as being
approachable, dealing personally with issues related to the school’s vision and by
personal example.
These results suggest that principals need to exercise a guardian role in terms
of protecting the vision of the school from distractions. Similarly Licata and Harper
(2001, p. 5) have argued that teachers who share in the vision “may be more likely to
implement their imagined possibilities and less likely to be distracted by some of the
more tedious routines and conflicts that tend to be part of everyday life in their
schools”. In the present investigation it was found that several teachers were unsure
as to the content of the vision. However, they relied upon their own personal
276
interpretations of what the school vision should be. These results imply that
principals need to ensure staff are fully aware of the school’s vision and its meaning.
This is particularly important given that previous research has found that teachers in
schools with poorly defined visions will often replace school vision with personal
vision (Vandenberghe, 1991).
Teacher “busy-ness” and lack of time for reflection can conspire to work
against the successful implementation of an educational vision. Several informants
recognised these two factors as possible threats to the perceived centrality of their
school’s vision. Other factors have also been identified in the literature as possible
distracters to school vision. Licata and Harper (2001), highlighting the difference
between leading and managing, indicate that even with thoughtful and caring
principals, schools may never know what could be achieved if they focus solely on
keeping things running smoothly. Many teachers in the present investigation reported
that an important way their school’s vision was implemented was through the daily
practices of the school. Although this suggested that staff see a positive relation
between their school’s vision and the daily practices of the schools, principals need
to be watchful that complacency does not dull a school’s vision.
Further, secondary school structures, as opposed to primary school structures,
which emphasise the “departmental differentiation of specialised teachers”
(Abolghasemi, McCormick & Conners, 1999, p. 81) can encourage balkanisation
among departments and faculties. Faculties may develop different subcultures that
act to reinforce certain teacher behaviours and faculty level policies, which are at
variance with school vision and policy. Among the staff interviewed, many saw this
balkanising effect as diluting the influence and centrality of their school’s vision.
Again, principals who recognise the mediating effect faculty heads can have, can
encourage a stronger sense of coupling among the various departments of their
school. This, in turn, will engender a greater commitment to the shared vision of their
school (Abolghasemi et al., 1999).
While the concept of vision was identified by respondents as a factor in their
respective principals’ leadership style behaviour, many participants reported being
committed to a personal vision rather than the actual vision of the principal or the
277
school’s vision statement as recorded. Factors ranging from lack of modelling on the
principal’s behalf, daily busyness and poor vision transmission were also found to
have the propensity to cause teachers to replace school vision with a personal vision
thereby eroding the effect of a common school vision.
Individualised Consideration. The second principal leadership factor
identified in Study 1 was individualised consideration. Each of the five items used in
the individualised consideration measure emphasised the leader’s behaviour of
spending professional time with each follower individually such as assisting,
coaching or otherwise engaging with the follower on a one to one basis (see
Appendix 5.1). Study 2 provided further support for the saliency of this construct
whereby teachers characterised the nature and usefulness of individual consideration
behaviours extended to them by their Principals. The results from Study 1 and Study
2 imply that individualised consideration is an important leadership style behaviour
that principals can exercise in schools because of its potential to enhance or erode the
relational nature of schools as communities.
The results from Study 1 also offer support for the suggestion that contingent
reward and individualised consideration act together, given that the individualised
consideration factor was derived from Bass’ (1997) individualised consideration and
contingent reward scales. These results also support those of Leithwood, Jantzi and
Steinbach (1999) who have argued that both contingent reward and individualised
consideration can mesh together, especially in public educational settings, where the
use of monetary rewards is not available. In their view, contingent reward can be
elevated to individualised consideration when the leader provides “informative
feedback about performance in order to enhance a teacher’s sense of professional
self-efficacy” (Leithwood et al., 1999, p. 73). In the present investigation,
opportunity for individualised attention from the principal to improve performance
was seen by participants as a motivator that raised teacher expectations about
reaching goals, and in turn contributed to a teacher’s overall sense of job satisfaction.
Therefore, the results of the present investigation suggest that individual
consideration can be utilised to transform multiple aspects of teacher behaviour.
278
The current investigation also resulted in identifying strategies implemented
by the principal that were perceived by teachers as indicative of individual
consideration. These strategies included: being seen around the school, taking an
interest in staff professional development and personal issues, and being seen to
“pitch in” to get tasks done. Similarly, principals also indicated that they considered
an important aspect of their job was to know their staff. Teachers emphasised that
they liked to be treated equitably and considerately by principals. Teachers also
reported that there were several “individualised support” principal practices that they
responded favourably to, including: not showing favouritism towards particular
groups or individuals, having an open door policy, being approachable and
welcoming, protecting teachers from excessive intrusions on their classroom work,
giving personal attention to seemingly neglected colleagues, and being thoughtful
regarding the personal needs of teachers. These practices were seen as encouraging
inclusion and community, and were specifically mentioned by many teachers as
strategies employed by their principals to get to know them. One strategy in
particular, known to staff as “the open door policy”, is noteworthy. Accessibility to
the principal, whether to discuss educational issues or personal matters, was valued
highly among many of the staff interviewed. Some staff saw this as their right rather
than a privilege accorded to them, while one principal interviewed believed it was
their duty to make time when staff wished to discuss matters. The use of an “open-
door” policy, which models both to staff and students an emphasis on maintaining
good interpersonal relationships, is one way that principal accessibility can be
enhanced (Dinham, Cairney, Craigie & Wilson, 1995). This particular practice also
was considered by teachers participating in the present investigation to have an
extremely positive influence on the creation of a positive school learning
environment. However, many teachers were careful in the application of this policy,
and reported that they preferred to arrange an appointment at an appropriate time.
Interestingly, it seemed as though knowledge of an open door policy was more
important than the actual use of it. These findings were consistent with those
reported by Sagor (1991), especially in terms of showing the need for strong care and
actively supporting the work of teachers, particularly in the area of teaching and
learning.
279
It also emerged from staff interviews that several strategies could be
employed to encourage individualised consideration within schools. These strategies
included: mentoring, delegation, and addressing specific teacher needs. Densten and
Sarros (1997) found that mentoring was an important sub-category of individualised
consideration, provided followers one on one contact with leaders, and gave
opportunity for meaningful and effective two-way communication to develop. Kirby,
King and Paradise (1992) similarly found that mentoring opportunities proved to be
an excellent substitute for monetary rewards, particularly in environments where
cash resources were in short supply. Staff reported that the principal mentored
teachers both formally and informally, networking with them, and by working
together on school projects. This finding is supported by Leithwood, Jantzi and
Steinbach (1999, p. 73), who also found that principals who “develop a close
knowledge of their individual colleagues” can often use these developed
relationships as the starting point for implementing school wide change.
The present investigation also found that individualised consideration can be
enhanced if principals use delegation strategies. Delegation can be defined as the act
of passing ownership for a decision to the followers, with the leader providing a
supportive role (Densten & Sarros, 1997). Participants in the present investigation
reported that being given authority and support in various aspects of school life (for
example, pastoral care, implementation of technology programs) also encouraged
them to greater commitment within their school communities. They also reported
that principals can encourage individualised consideration by giving attention to the
specific needs of individual staff. They suggested that this might include representing
a staff member to a higher authority, such as the Department of Education and
Training, or arranging time off in the case of a personal emergency. Participants also
suggested that “modelling” was a powerful tool that could be used to promote
individualised consideration. Consistent with previous research, it was also identified
that: (a) modelling occurred when the leader “acts as a role model, leading by doing
rather than only by telling” (Leithwood, Jantzi & Steinbach, 1999, p. 80); (b)
modelling was seen as setting a clear example that staff could follow that was
consistent with the values and attitudes the leader espoused (Kirby et al., 1992), and
(c) modelling recast a commitment to the school, professional growth and the process
of group problem solving (Leithwood et al., 1999). One teacher, when considering
280
the principal’s modelling approach to leadership, commented that he was even
willing to do substitute teacher duties (extras), not only to assist an absent staff
member, but also to talk with students and get a feel for their thinking on different
issues. An important finding to emanate from this investigation was the influence of
the principal, in the act of “rolling up their sleeves” (Leithwood et al., 1999, p. 81)
and involving themselves in the daily business of education, sending a persuasive
message that promotes individualised consideration within their schools.
The last finding reported by teachers that emerged from an analysis of the
data was the role that recognition played in the promotion of individualised
consideration. This recognition took both private and public forms, and occurred
when principals recognised good work and effort (Leithwood et al., 1999) by giving
individual praise, speaking specifically about what constitutes good work, offering
personal encouragement to individuals for effort and performance and demonstrating
confidence in teachers’ ability to perform to a high standard. Sagor (1991) describes
this recognition strategy as either grandstanding (for example, taking the floor to
expound a position which emphasises a core value) or cheerleading (for example,
telling a visitor about individual or team accomplishments within the school), and
notes that often times it is done with humour and much laughter. Furthermore, many
of the teachers interviewed could readily identify several forms of recognition
strategy used in their schools. The latter included individual recognition, where the
principal personally tells the staff member directly that they have done good work,
and group or public recognition, where formal or informal events such as assemblies
are used to “grandstand”.
In summary, whilst the quantitative phase (Study 1) of the present
investigation identified individualised consideration as a leadership factor, the
qualitative study enriched the findings by elucidating how these behaviours were
operationalised in practice. It was found that principals who modelled individualised
consideration strategies set and influenced the relational standards and behaviour
followed by staff and students in their respective communities. Further, these
principals also emphasised the teaching and learning function of their schools as core
business. This was achieved by employing strategies from implementing an open
281
door policy through to other inclusive strategies such as recognising effort and
achievement in all its forms across the broad spectrum of school life.
Laissez-Faire Leadership. A third leadership style behaviour, laissez-faire
leadership, was identified as a result of Study 1. These results support those of Bass
and Avolio (1997) and indicate this type of leadership style behaviour is present in
Australian schools. Whilst qualitative data was not specifically collected in relation
to laissez-faire leadership behaviours of school principals given participants in Study
2 believed their principals demonstrated few or no laissez-faire leadership style
behaviours, respondents did offer insight into how non-laissez-faire principals dealt
with the decision-making processes in their respective schools (Table 6.18). Three
leadership practices were identified by respondents as important in the decision
making process. These practices included the recognition that the principal had the
final right of veto in the decision-making process, even when other staff members
were involved. These practices also included staff participating in decision-making
within their schools by standing on executive committees, and recognising the
importance of the chain of command (that is, the executive school management
structure) as conduits in the decision-making process.
Summary Aim 1: Differential Leadership Attributes. While evidence for
Bass et al.’s (1997) operationalisation of the transformational and transactional
leadership paradigm could not be completely supported, certainly the results
emanating from both quantitative and qualitative studies suggest that a visionary
leadership construct (transformational leadership), an individualised consideration
construct (a hybrid transformational/transactional factor), and laissez-faire leadership
(a non-leadership construct) could be identified within a sample of New South Wales
secondary schools.
Discussion of Aim 2: The Differential Relation of Leadership Styles on School Learning Environment
The second aim of this study was to test the differential impact of different
leadership styles of secondary school principals on selected school learning
environment constructs. Study 1 found that the three leadership styles identified had
282
a differential influence on four of the seven school learning environment constructs
examined in this study: student supportiveness, affiliation, resource adequacy,
achievement orientation. Further, analysis showed that the influence of these
leadership style behaviours on the four school learning environment factors was
different at the two levels of analysis, namely, at the school level and teacher level
(for example, Tables 5.1 and 5.2). This is an important consideration, as the
influence of leadership style is expected at the teacher level as different teachers
respond to the leadership behaviours used by their principals. The influence of
principals’ leadership behaviours at the school level is suggestive of the potential
these strategies have to affect an entire school community, and potential potency as
far as school wide change is concerned.
Affiliation shared a statistically significant relation with the leadership
construct of individualised consideration. This finding indicates that staff perception
of affiliation or collegiality is positively influenced by perceptions of individualised
consideration leadership behaviour (Table 5.2). Study 2 also gave insight into how
this may occur. Respondents indicated that the principal could apply several
strategies that were perceived as being conducive to fostering a sense of affiliation
(Table 6.9). These included strategies such as the principal acting in a genuine and
sensitive manner towards staff, and showing care and concern for staff in both
professional and personal areas. Other strategies included: making time available for
staff to discuss professional and personal matters, spending time at school events,
and acknowledging staff effort and involvement in the life of the school. As
suggested earlier, these results imply that the modelling behaviour of principals is a
powerful strategy that can be used to promote affiliation (Leithwood, Jantzi &
Steinbach, 1999).
Statistical analysis of the data collected indicated that the laissez-faire
leadership style of principals at the school level was significantly and positively
related to perceptions of teacher affiliation. If staff respondents perceived high levels
of laissez-faire leadership style behaviour displayed by their principal, then their
perceptions of affiliation or staff collegiality also rose. Perhaps when leadership is
absent or avoided by the principal, staff will band together collegially in strong sub-
units or groups. Previous research has found that a leadership substitution effect
283
(Dionne, Yammarino, Atwater & James, 2002; Kerr & Jermier, 1978; Swenson,
2000) occurs when leadership seems irrelevant to an organisation, or when
leadership becomes dysfunctional, and that schools are particularly prone to a
substitutionary leadership effect (Swenson, 2000). Teachers are highly trained, and
work in small groups that rely less on supervision, and where feedback and collegial
support is available from peers. It seems reasonable, therefore, that as laissez-faire
leadership style behaviour is perceived to increase, so too does teacher perception of
affiliation. This style of leadership may foster collegial relations to the point where
no one group member’s decision-making is considered more important than
another’s, and so a genuine atmosphere of working together is created.
Resource adequacy was the only school learning environment factor to share
a statistically significant, negative relation with principals’ visionary leadership
behaviours (Table 5.6). This result suggests that when teachers’ perception of
visionary behaviour increases, teachers’ perception of the adequacy of teaching
resources decreases. Examination of the qualitative results yielded further insights
into this relation. Teachers perceived that resourcing for schools is locked into some
funding formula that is administered by a central educational authority, and that
resources are shared out according to need (Table 6.17). Generally, classrooms are
considered run down (Table 6.15) and resources are in short supply (Table 6.16). So
if a principal has an idea regarding the implementation of a new program (visionary
leadership behaviour), staff are often expected to make existing resources spread to
cover this new program. Although it is acknowledged that the principal will, in many
cases, follow the advice of the school executive (Table 6.18) in purchasing
educational resources, even programs aimed at upgrading the school’s physical
environment will require the channelling of funds away from teaching resources.
Some respondents went so far as to suggest that the whole education system is
running down (Table 6.16), and that teacher morale was at a low point because of
this. This being the case, perhaps new initiatives emanating from visionary
leadership behaviours would be perceived by teachers as impacting on the already
restricted resources available to fund existing programs. It is also interesting to note
that in addition to affiliation and resource adequacy, a statistically significant,
negative relation was present between laissez-faire leadership behaviour and
teachers’ perceptions of resource adequacy (Table 5.6). This suggests that teachers
284
perceive an increase in leadership avoidance behaviours is associated with a decrease
in the availability of resources. Perhaps teachers perceive that laissez-faire principals
cannot distribute or manage teaching resources equitably.
Laissez-faire leadership behaviours were also found to share a statistically
significant, positive relation with teachers’ perceptions of both student
supportiveness and achievement orientation (Tables 5.2 and 5.8). These results
suggest that teachers perceive that increases in laissez-faire leadership behaviours in
their principals are related to positive teachers’ perceptions of student supportiveness
and achievement orientation. However, this result might echo the effect that
substitute leaders can have (e.g. autonomous staff such as teachers who work in
teams), and as such the result is likely to be confounded. For example, collegial work
groups who emphasise the relational aspects of school life may well counter the
effects of a principal who is perceived to be laissez-faire in the relational aspects of
school community life. Hence, this result warrants further investigation.
Finally, three of the seven school learning environment measures
(professional interest, centralisation and innovation) recorded no statistically
significant relation with any of the three leadership styles under consideration in this
study. This suggests that principals may have a limited influence on these aspects of
learning environment and must look to other means apart from the application of
leadership style to do so.
Summary Aim 2: Differential Relation on School Learning Environment.
In summary the results of the present investigation demonstrate that principals’
leadership style behaviours have a differential impact on school learning
environment. Surprisingly, while visionary and individualised consideration
leadership behaviours were demonstrated to have a limited influence on school
learning environment constructs, laissez-faire style behaviours were shown to impact
positively upon student supportiveness, affiliation and achievement orientation, and
negatively on resource adequacy. These results suggest that it might be useful for
future research to identify what facets of laissez-faire leadership style might be
conducive to influencing school learning environment constructs.
285
Discussion of Aim 3: The Differential Relation of Leadership Styles on
Perceptions of Satisfaction, Influence, Effectiveness and Teacher Control
The third aim of this study was to test the differential impact of different
types of leadership styles of secondary principals on teachers’ self-perceptions of
satisfaction with their principal, and teachers’ self-perceived effectiveness to produce
desirable educational outcomes. An analysis of the data collected as part of the
quantitative phase of this study indicated that the three leadership styles of visionary
leadership, individualised consideration and laissez-faire leadership exerted a
differential relation with each of the teacher outcome measures under examination
(see Table 5.13). The most surprising and controversial finding was that visionary
leadership style behaviours of principals were indicated as having no statistically
significant relation with teachers’ perceptions of global satisfaction with leadership,
influence, effectiveness and control. This is a startling result in that it is contrary to
findings reported in other industries (Avolio, Bass & Jung, 1995), and contrary to the
theoretical underpinnings of the transformational and transactional leadership
paradigm (Bass, 1985). However, the remaining two leadership style behaviours,
individualised consideration and laissez-faire leadership, did indicate a statistically
significant relation with teachers’ perceptions of global satisfaction with leadership
and perceptions of influence. Individualised consideration demonstrated a strong,
positive, statistically significant relation with teachers’ perceptions of global
satisfaction with leadership and a negative but statistically significant relation to
teachers’ perceptions of control. These results suggest that both teachers’ perceptions
of satisfaction and control over their own workloads are significantly related to the
leadership style exercised by their principals. By contrast, laissez-faire leadership,
unsurprisingly, indicated a negative, statistically significant relation with both
teachers’ perceptions of global satisfaction of leadership and perceptions of
influence. This result also reinforces the finding that teachers like to be led by a
principal who affirms them and knows and cares for them as individuals, but not by a
principal who avoids decision-making and is absent when important issues arise.
Overall these results imply that teachers are influenced more by their
principal’s individualised consideration behaviours than they are by visionary
leadership behaviours. Further, and as expected, teachers’ perceptions of global
286
satisfaction are negatively influenced by their principal’s laissez-faire leadership
behaviour. The results indicate that teachers’ scores for global satisfaction with
leadership are higher when related to principals’ individualised consideration
behaviours than they are for principals’ visionary or laissez-faire leadership
behaviours. This suggests that teachers prefer and are satisfied with, principals who
know them individually rather than principals who are absent when needed.
An analysis of the data gathered also indicated that a principal’s
individualised consideration behaviour negatively influenced teachers’ perceptions of
influence (Table 5.13). Although the reason for this is unclear and speculative, it
suggests that teachers’ perceptions of their influence are diminished as principals
demonstrate individualised consideration behaviours in their communities. This may
well be because as principals interact with their staff and get to know their strengths
and weaknesses, they can implement professional development, teaching
assignments and extra curricular duties to their own design at the expense of what
individual teachers desire, diminishing teachers’ sense of being able to exert
influence in their schools. This is an area that needs further investigation.
In this current study, several intrinsic motivators or satisfiers were identified
as a result of the qualitative survey. Most of these motivators emphasised the
relational nature of teaching as a profession. Motivators included factors such as
receiving encouragement, both as a group and individually, acknowledgement of
effort, the ability to develop strong relationships with students, and the positive
influence of working in a collegial environment that is free from unpleasant clashes.
These factors, according to the informants’ reports, were those that were satisfying as
far as their jobs were concerned. An important satisfier, at least in the perception of
some of the staff interviewed, was the modelling role the principal played in the daily
life of the school. Several staff felt that because the principal was seen associating
with them in daily school life, following the principal’s lead was an easy and
satisfying thing to do. These results are consistent with findings by Bogler (2001, p.
679) who has argued that it is through this participative behaviour that “principals
can develop and foster positive feelings and attitudes of teachers regarding their
vocation”. Staff in the second phase of the study identified several possible factors
that operated as satisfiers within their schools. These school-level factors included
287
quality and clarity of communications, flexible application of school rules and
regulations, the positive use of the “chain of command” executive structure within
the school, and the perception of the school as a happy environment. Each of these
factors has the potential to act in a dissatisfying manner, adversely influencing
teachers’ attitudes towards their occupation. Conversely, these factors can also make
attending to the daily job of teaching, a satisfying vocation.
Several dissatisfiers were also elucidated in Study 2. Most of these
dissatisfiers were extrinsic to teachers, in that their effect was extraneous and came
from outside the locus of control of the individual teacher. Factors included in this
domain related to the task of teaching, and were reported in terms of increased
workloads, expectations of doing more with fewer resources, stress from the pace of
educational change, low morale and low public esteem of teaching as a profession.
One interesting dissatisfier to emerge was the juxtaposition of enjoying teaching but
having a general feeling of animosity towards the structures of teaching imposed by
the Department of Education and Training. These results are supported by Dinham
and Scott (1997), who consider this domain, that is, the area between the intrinsic
satisfiers of teaching and the extrinsic dissatisfiers of educational change, workload,
and structures as the most important domain of all. This domain is made up of school
based factors including: school leadership, school learning environment and school
reputation, and is the domain over which schools and principals have most control. It
appears that the factors within this domain have the propensity to act either as
motivators or dissatisfiers in schools, depending on how they operate within their
environments.
Summary Aim 3: The Differential Relation of Leadership Styles on
Satisfaction, Influence, Effectiveness and Teacher Control. In summary the results
of the present investigation indicate that leadership behaviours of school principals
did demonstrate a differential relation to the various teacher outcome measures
considered in this study. However, the lack of influence of the transformational
leadership style of visionary leadership was a major finding that has important
implications for an understanding of the role of vision in schools as a source of
teacher satisfaction. Teachers, it would seem, are more satisfied by the individualised
288
consideration behaviours of their principals than they are by their principal’s
visionary leadership style behaviours.
Discussion of Aim 4: The Differential Relation of Antecedent Variables on
Leadership Styles and School Learning Environment
The fourth aim of the study was to examine the relation between different
types of leadership style of secondary principals, selected school learning
environment constructs and selected school and teacher level antecedent variables. A
complex relation was found to exist between the antecedent variables examined in
the study, leadership style behaviours, and school learning environment.
Furthermore, this analysis revealed that differential relations were evident at both the
school level and the teacher level.
School size. The antecedent variable of school size is related to the number of
students in a school. At the school level, school size was found to be the only
antecedent variable to have a statistically significant negative effect on teachers’
perception of the individualised consideration behaviours of their principals (Table
5.17). Whereas school size was not mentioned specifically by respondents in
interview, several made inferential references to the principal knowing both the staff
(Tables 6.8 and 6.9) and the students (Table 6.11) as something easier to maintain
when a school is smaller. Individualised consideration, as indicated earlier, had in
turn a statistically significant, positive relation with teachers’ perceptions of student
supportiveness. These results imply that there is an optimal school size before the
numbers of students means that maintaining individual relations with students
becomes difficult. Cotton (1996), in reviewing the research on school size, school
climate and student performance in the United States up to the mid 1990’s suggested
that small schools (in terms of student population) show better, statistically
significant results than do larger schools in terms of student absentee rates, dropout
rates, student sense of belonging, self-concept, and interpersonal relationships. The
results of the present investigation suggest that there is an optimal school size in
terms of student population. After this size is reached, then the principal’s
effectiveness in terms of influencing teachers’ perceptions of individualised
consideration diminishes.
289
School size at the school level was, however, indicated to have a statistically
significant, positive relation with resource adequacy. This indicates that teachers’
perception is that the larger the school in terms of student population, the more
access to teaching resources they have. To a certain degree this perception may be
true, given the central office practice of tying funding to student numbers.
Teacher position. The antecedent variable of teacher position refers to
whether staff held teaching or administrative positions within schools. At the school-
average level, teacher position in schools was demonstrated as having a statistically
significant, negative relation on perceptions of achievement orientation (Table 5.17).
This may simply indicate the perception among teachers that co-ordinators’
workload in schools is such that they do not have the time to emphasise an
achievement orientation in their classes.
Teacher position at the teacher level was indicated as having a negative,
statistically significant relation with each of: teachers’ perceptions of principal’s
vision, student supportiveness, centralisation, innovation and resource adequacy. As
staff rose from purely teaching to administrative and executive positions within
schools, teachers perceived that each of these leadership and school learning
environment factors decreased. This could be interpreted as perhaps reflecting the
cynicism that established teachers may feel the longer they stay in the one school.
Further, perhaps cynicism intensifies, the longer an executive teacher serves with a
principal in the same school. The practice of teachers moving between schools after a
number of years of service in one particular school may therefore be useful, as it
provides teachers with a new context.
Teacher time in current school. The antecedent variable of teacher time in
current school simply refers to the number of years a teacher has served in one
school. Teacher time in current school at the teacher level was demonstrated to have
a statistically significant, negative relation on teachers’ perceptions of individualised
consideration, and a statistically significant positive relation on teachers’ perceptions
of laissez-faire leadership. Teachers who have been in their schools for a long time
are less affected by their principal’s individualised consideration behaviours. Further,
these teachers also respond well to their principal’s laissez-faire leadership
290
behaviours. Again, this may simply reflect teacher cynicism, the longer a teacher
serves with a principal. Alternatively, these teachers may feel confident and
experienced enough in what they do to need little support. This is consistent with
some of the findings arising from substitute leadership studies within educational
settings (Dionne, Yammarino, Atwater & James, 2002).
Teacher time in their current school at the teacher level was also indicated as
having a statistically significant, negative relation to centralisation. This suggests that
teachers perceive the longer they remain in the one school, the less able they are to
influence the decision-making process. This may well be a behavioural response to
previous decision-making processes that have not produced the desired results.
Alternatively, it may signal that longer serving teachers are not so concerned with
decision-making and school management issues, but would rather focus on other
concerns, for example, teaching and learning processes.
Teacher gender. The antecedent variable of teacher gender indicated a
statistically significant positive relation with teachers’ perceptions of principal’s
vision. This gender effect suggests that female teachers are more receptive to
visionary leadership behaviour than are male teachers. Teacher gender was also
positively related to teachers’ perceptions of achievement orientation. This indicates
that female staff are also perceived by staff generally as being more inclined to push
students to higher levels of achievement within their studies.
Time spent in a school with current principal. This antecedent variable
refers to the amount of time a teacher has served in the one school with the same
principal. This result was similar to other “experiential” type variables (for example,
teacher experience and teacher time in current school), in that the results
demonstrated that teachers who had served a long time in the same school with the
same principal were less inclined to perceive their principal’s visionary behaviours
than were teachers who had served in the one school with the same principal for
shorter time periods. Again, issues of teacher autonomy may well be influencing this
relation, and a policy of encouraging movement between schools could alleviate this.
291
Principal gender. The antecedent variable of principal’s gender at the school
level was demonstrated as having a statistically significant, positive relation with
innovation, and a statistically significant, negative relation with centralisation (Table
5.17). On one hand, male principals are perceived by staff to centralise or focus the
decision-making processes in their schools, usually into a senior executive, while
female principals are perceived to be more adaptive to innovative change.
Summary Aim 4: The Differential Relation of Antecedent Variables. In
summary, at the school level it was not surprising that teachers perceived a negative
relation between the number of students on a campus and the level to which they
perceived their principal’s individualised consideration behaviours. At the teacher
level, variables related to experience and mastery of teaching (position, time in
school and time in school with current principal) demonstrated a negative influence
on perceptions of the principal’s leadership behaviours, and as such perhaps related
to teacher maturity rather than any motivational influences the principal’s leadership
style may have had. Further, it was also expected that school size would show a
strong relation with teachers’ perceptions of resource availability in schools.
The study also found some gender differences in teachers’ perceptions of
school learning environment and teacher outcome measures. At the school level,
gender differences were evident in teachers’ perceptions of the level of student
supportiveness. At the teacher level, gender differences in perceptions of
achievement orientation, teacher influence and individual teacher perception of
visionary leadership behaviour also became apparent.
Discussion of Aim 5: Identify leadership strategies that enhance or erode teachers’
perceptions of school learning environment and teacher satisfaction
The fifth aim of this investigation was to identify leadership strategies that
teachers perceive as either enhancing or eroding the school learning environment and
teacher satisfaction.
Enhancing perception of school environment. Teachers identified the
celebration of success, however small and from whatever source, as an inclusive and
292
unifying factor throughout their school communities. Teachers viewed principals
who regularly and publicly celebrated success as enhancing their school’s learning
environment, and as creating a milieu where more success was anticipated, expected,
desirable and possible (Starrett, 1991; Table 6.12). Another central tenet to the
enhancement of a positive learning environment expressed by teachers was the need
to make schools safe and secure places; teachers suggested a number of strategies
that could be employed (e.g. safety audits, name badges) to address student safety
(Table 6.15).
The degree of professional trust between teachers was also identified as an
important factor in enhancing a school’s learning environment. Staff alluded to
several indicators of the level of professional trust (Table 6.15). Comments that
highlighted a reliance on peers to competently attend to their jobs were common, as
were comments regarding the positive input staff shared with each other. Principals
were seen by teachers to positively influence the morale and social climate in their
schools by taking a personal and professional interest in their staff members, and
modelling the type of behaviour that teachers demonstrated towards each other and
their students (Table 6.16).
Teachers indicated that a positive school learning environment was one
where teaching and learning activities are emphasised as the core business of their
schools. Teachers stressed that classes should be taught well, implying that lessons
must be carefully planned and meaningful (Table 6.15). They also suggested that it
was important to hold expectations that all students can be successful in learning, and
vital to customise the curriculum to meet individual needs, along with allocating
quality resources to ensure curriculum could be effectively taught (Table 6.17).
The principal’s encouragement to innovation in teaching was also viewed by
participants as having the potential to enhance a school’s learning environment
(Table 6.15). However, the corollary of this, at least in the perception of some of the
teachers interviewed, was that the speed of implementation of innovation could be
detrimental to the development of a school’s learning environment (Table 6.16).
Most staff interviewed readily embraced innovation and change as a part of the
teaching process (Table 6.18). However, a few commented that some colleagues
293
found change difficult to accept, that some staff were suspicious and cautious about
implementing any innovation, and therefore some needed to be coerced (Table
6.18). So, while any innovation that assists teachers in the classroom is welcome, the
rapidity at which innovation is implemented causes some teachers undue stress. It is
the role of the principal to guard the integrity of a school’s learning environment by
carefully facilitating and managing the speed at which innovation is implemented
(Hall & George, 1999). An interesting contrast can be drawn between the role of the
Department of School Education, as the source of much change in the New South
Wales educational system, and the school principal, who acts as their agent. Many
teachers saw change as inevitable, but became disgruntled at the system
administrator rather than at the school administrator. On the one hand, teachers
reported disdain for the system (Table 6.16), but at the same time, commitment for
the job and support for the principal, who was also perceived to be under the control
of the system.
Teachers reflected the positive influence that communication systems had on
their daily activities (Tables 6.6). Principals encouraged open lines of
communications in several ways, including speaking directly to staff about
educational issues, being seen around the school and utilising an executive structure
to funnel information to and from staff. Participants also recognised that they could
have a role in the decision-making processes in their schools. Often times this was
done formally through the executive structure, where opinions were deliberately
solicited and reported back through a chain of command. Opportunity was also given
through open staff meetings or through surveys. Interestingly, staff did not object to
the fact that the principal had the final right of veto in the decision-making process. It
seemed that a positive enhancement to school learning environment came from
involvement in the decision-making process, rather than in forcing one particular
point of view forward over another. Conley and Muncey (1999) reported similar
findings regarding the effectiveness of teacher involvement in the decision-making
process from research carried out in the United States.
Many teachers surveyed suggested that the notion of relativity was important
when comparing the resources of their schools with those of other schools (Table
6.18). Although funds are always in short supply, informants’ schools were
294
inevitably contrasted with other local schools. This perception positively impacted
upon the attitude of teachers, and hence had a positive impact on the school’s
learning environment as teachers in the present investigation perceived their schools
to hold more resources in contrast to other local schools.
The physical state of buildings and grounds was found to influence the
learning environment of a school (Table 6.17) positively. Teachers and principals
also made several references to the impact that the physical school buildings can
have on the learning environment. Working in a well-presented physical environment
had a positive influence on teachers’ attitudes and perceptions of the value of being
employed in their schools. As well, principals demonstrated the value they placed on
providing a positive physical environment by allocating funds to improve and
maintain buildings and grounds, appointing specialised positions within their
schools, and being involved themselves with school renewal programs. Mok and
Flynn (1997) found similar results when surveying student preferences from schools
in Sydney’s western suburbs, as did Stanton (1999) and Duke (1998) when surveying
schools in rural areas of the United States of America.
Eroding perceptions of school learning environment. Teachers indicated
several factors they saw as being detrimental to creating a positive school learning
environment within their schools. Some of these factors were related to the school’s
vision, personal relationships, the physical state of available resources, and
community perceptions of teachers.
Teachers reported several key aspects regarding the development and
implementation of vision that had the potential to erode teachers’ perceptions of
school learning environment. Factors ranging from not clearly defining vision and
not regularly referring to the school’s vision, to not allowing new staff to have an
input into the formulation of the vision (Table 6.4) had the potential of affecting a
teacher’s perception of their school learning environment. One key element
mentioned by teachers was the effect that the daily grind of teaching had in diluting
the importance of the school’s vision (Table 6.7). This needed to be guarded against,
as it allowed the possibility of replacing school vision with personal vision, and
disaffecting teachers’ perceptions of school learning environment.
295
Teachers’ perceptions were that schools existed to provide a positive
educational experience for the students in their care (Table 6.5). The majority of staff
believed that this could be done through creating and maintaining positive personal
relationships with their students (Table 6.14). Anything, therefore, that detracted
from developing the relationship teachers had with principals, other teachers and
students, had the potential to erode teachers’ perceptions of their school’s learning
environment. As far as students were concerned, poor welfare and discipline systems
(Table 6.17) had the potential to erode teachers’ perceptions of their school’s
learning environment, and not giving students authentic power through channels such
as the Student Representative Council also had the potential to erode teacher
perceptions of school learning environment.
Teachers also indicated that teacher-to-teacher relationships were important
in maintaining positive school learning environments (Table 6.14). Those aspects of
collegial relationships that had the potential to dissatisfy teachers and erode
perceptions of school learning environment included lack of teamwork, and not
being able to rely on other teachers to do their jobs properly (Table 6.16).
Teachers believed that at least some of their principals’ leadership
behaviours had a large and positive impact on the creation of a positive school
learning environment. Presumably, these behaviours when missing or absent would
have a negative effect on teachers’ perceptions of school learning environment.
Unclear lines of communication would have the effect of clouding issues and
distorting priorities, while not considering teachers’ viewpoints in the decision-
making process (Table 6.6) would isolate teachers’ opinions from principals. Further,
principals who stayed secluded and unavailable to their staff also risked eroding staff
perception of a positive school learning environment (Table 6.10).
Teachers suggested several factors, related to physical resources available for
teaching, that have the potential to erode perceptions of school learning environment.
Some teachers commented that there were fewer teaching resources available to do
the work required (Table 6.16), while others expressed a desire to make classrooms
educationally stimulating (Table 6.15). Presumably, the absence or lack of resources,
296
or buildings that were physically dilapidated and run down all worked against
creating a positive perception of a school learning environment.
Some teachers indicated that they felt the general community’s perception of
teachers was fairly low, and that teachers were not held in high esteem within the
respective communities. The effect was that teacher self-esteem was low, and that
teachers were burning out as a result (Table 6.16). Several staff across the schools
surveyed responded to this by indicating while they like the school they were in, they
disliked the education system as whole.
Summary Aim 5: Identify leadership strategies that enhance or erode
teachers’ perception of school learning environment and teacher satisfaction. In
summary, the findings of the present investigation suggest that a principal’s
leadership style has the potential to either erode or enhance teacher perceptions of
school learning environment and teacher satisfaction. Participants identified several
behaviours and strategies principals can employ that will enhance teacher perception
of school learning environment. It was found that celebrating success and enhancing
the quality of professional relationships between staff promotes a positive school
environment. Results of the present investigation also indicate that principals can
manipulate administrative strategies and practices that can enhance a school’s
learning environment. For example, teachers emphasised that they appreciate being
consulted in the decision-making process. Finally, the physical state of the buildings,
grounds and classrooms was also found to positively influence the learning
environment of a school. Conversely, factors such as allowing buildings and teaching
resources to run down had the potential to erode teachers’ perceptions of school
learning environment, as did a general despondency of working within an
educational system like the Department of Education and Training.
Strengths and Limitations of the Study
This present study offers a number of strengths in comparison to other recent
school-based leadership research. Central to this has been an examination of the
intuitively related, but largely unresearched concepts of principal leadership style
297
behaviours and their differential impact on school learning environment and selected
teacher outcomes. The application of Bass et al.’s (1997) conceptualisation in an
Australian educational context and the utilisation of a multilevel modelling
methodology also is novel.
The present investigation also has a number of key methodological strengths.
Firstly, this study utilised a large and diverse sample of secondary teachers and
schools across the state of New South Wales, Australia, so that sophisticated
multilevel statistical modelling techniques could be employed. The use of multilevel
modelling techniques enabled the consideration and elucidation of multiple effects at
both the school and the teacher levels. This current study also examined the
inherently group level constructs of leadership and school learning environment.
Methodologically, this means that individual perceptions of these constructs may
well be different to group level perceptions of the same constructs (Papaioannou,
Marsh & Theodorakis, 2003). Any research that fails to account for the multilevel
nature of leadership and school learning environment constructs will produce at best
dubious results. Previous research has had limited possibilities because of limitations
in existing statistical methodology, whereas the introduction of multilevel statistical
analyses should have a profound impact on this area of research. However, by
specifically employing a multilevel model approach in analysing the survey data,
variance at both teacher and school level was accounted for in the data. Another
methodological strength of the present investigation was the utilisation of a
synergistic blend of quantitative and qualitative research methodology which enabled
the rich tapestry of relations between constructs to be more fully examined than what
would be possible using one type of research methodology alone.
While this study was innovative in its approach to conceptualising school
leadership and its application of multilevel modelling techniques, several study
limitations need to be considered when interpreting the findings. The statistical
analyses employed in Study 1 relied on data obtained from a single source. While
single source approaches are common in leadership studies, it is recognised that such
an approach has the potential to bias any relations that are supported in the data.
298
A further limitation of Study 1 relates to the issue of missing data. It was
apparent by examining the returned questionnaires that some teachers tired of
answering questions, simply leaving the tail end of the instrument blank. Removing
cases with more than twenty percent missing data reduced the overall number of
available cases for analysis, and the further imputation of values (using the EM
method) for the remaining missing data could possibly have some effect on the
results. Study 1 was also limited by the use of the Multifactor Leadership
Questionnaire (MLQ-5X Short; Bass & Avolio, 1997) as a measurement instrument.
The instrument failed to discriminate the full range of transformational and
transactional constructs purported in the literature (Bass & Avolio, 1997). Future
researchers may well need to develop psychometrically sound measurement
instruments that clearly discriminate between specific leadership styles hypothesised
in the literature.
In relation to Study 2, the limited sample size may limit the generalisability
of the findings. Finally the findings reported in this study may only be generalisable
to government sector secondary schools in New South Wales. While several of the
findings and strategies suggested by this study are universal in nature and therefore
transferable to other non-government educational sectors, further research would be
needed to verify these findings in other contexts.
Implications for Research and Theory in Educational Leadership
This investigation has afforded insight into the effectiveness of different
principal leadership style behaviours in selected secondary schools across New South
Wales, and their associated impact on perceptions of school learning environment
and teacher outcome measures. The findings contained in this study also have the
potential to act as a catalyst to direct further research in this area in a number of
ways. The results of the present investigation imply that the conceptualisation or the
operationalisation of the transformational and transactional leadership paradigm
(Bass, 1985; Bass & Avolio, 1997; Burns, 1978) is weak. As such it is disappointing
to note that results reported from other applications (military and business) have not
been reproduced in the education sector in New South Wales, Australia. Teacher
299
reports in the present investigation demonstrated that transformational leadership
does indeed exist in Australian schools, as indicated by both the extraordinary
commitments to school vision and mission by some staff and the results obtained by
schools in terms of student achievement. More consideration needs to be given to the
conceptualisation and operationalisation of the transformational and transactional
leadership paradigm. Importantly, there is a dire need to develop psychometrically
sound measurement instruments that measure theoretical conceptualisations of
leadership behaviour. Given the current worldwide trend of economic rationalism
applied to education, such research seems timely and has the potential to inform
principal training and professional development and enhance effective schooling.
Athough the link between school leadership and school learning environment
has been reported as both intuitive and tentative (Griffith, 1999), it is nonetheless real
and has the potential to be far reaching in effects on both teacher and student
outcomes. Further, it needs to be recognised that learning environment effects are
both multifaceted and multilevel. While this study has considered some of the more
obvious school learning environment factors, future research could also benefit from
examining other influences such as parents and communities. As well, the effects of
school learning environment, at classroom, at school, at district and even at state
levels need to be accounted for. As communities demand increasing accountability
and higher results in terms of teacher and student outcomes, it makes sense that
effective school learning environment factors would be maximised so as to achieve
the best possible outcomes for all stakeholders. Lastly, given the current worldwide
shortage of teachers, and problems with declining retention rates among qualified
staff, any program aimed at fostering increased teacher satisfaction has to be judged
as worthy. Again, in an economic rationalist educational environment, if teachers’
perceptions of job satisfaction can be increased without the need to spend large
amounts of money, this must be considered worthwhile.
Summary
Overall, these results have indicated the differential influence that principal
leadership behaviours can have on selected school learning environment factors and
300
teacher outcome measures. The results demonstrate the link between school
leadership, school learning environment and teacher outcomes, and are suggestive of
the means whereby the principal’s leadership behaviour may be diffused throughout
schools. In this chapter it has been demonstrated that the results achieved were based
on a carefully constructed research design aimed at avoiding some of the
methodological problems encountered in previous leadership research. Further, a
sophisticated and appropriate multilevel statistical technique has been used to test the
hypotheses proposed, and a rigorous and proven qualitative approach has been
utilised to consider the research questions raised.
301
CHAPTER 8
SUMMARY AND CONCLUSION
Much has been written in the effective schools literature about the influential
role the principal plays in setting and enhancing the educational goals and agenda for
their learning communities. Furthermore, this literature has suggested that the link
between what principals do, school learning environments and teacher outcomes is a
rich and potentially fertile area for educational research. Unfortunately, the paucity
of empirical evidence in this area has meant much of this research has relied on
intuition rather than empirical research. Thus, one of the goals of the present
investigation was to examine empirical evidence linking principals’ leadership style,
school learning environment, and teacher outcomes. Furthermore, the study was
devised to address a number of timely issues in relation to effective schools. Firstly,
in an age of economic rationalism applied to public sector utilities such as schools,
there has been an international trend to increase the accountability of educational
administrators and a push to achieve more, as far as teacher outcomes are concerned,
with less expenditure. Secondly, new paradigms of leadership have recently emerged
that have offered the promise of organisational performance beyond what can
reasonably be expected with the application of particular leadership styles. This
transformational/transactional conceptualisation of leadership has provided the
theoretical framework that has guided this inquiry, especially in the area of the role
of school learning environment as a transmission mechanism between leadership and
teacher outcomes. This is a particularly attractive area, given the often-assumed
propensity of transformational leaders to modify their environments rather than
simply react to them.
The present study has contributed significant gains to an understanding of the
relation between principals’ leadership behaviours, school learning environment and
teacher outcomes, and has provided a solid, empirical foundation from which to
understand the transmission lines between these constructs. This was achieved by
employing a statistically sensitive multilevel modelling technique designed to tease
out the sources of variation at both the teacher and the school level.
302
This study has also broken new methodological and substantive grounds on
which previous research has either failed or been unable to capitalise on. Previous
research up until this point in time has had limited success because of a failure to
take into consideration the nested nature of school data. It is also important to note
that respondents who share the same school experiences together will likely respond
in a similar manner to survey instruments. This is particularly crucial in this study as
both perceptions of leadership and perceptions of school learning environments are
group constructs; that is, to be properly understood, they must considered at both an
individual level and a group level. Thus, the multilevel modelling technique
employed in this study provides the statistical tools from which to appropriately
assess these individual level and group level constructs. Furthermore, substantive
issues related to the impact of principal’s leadership style behaviours on teacher’s
perceptions of school learning environment and selected teacher outcomes have also
been addressed. Study 1 has made significant contributions to an understanding of
principals’ leadership roles in New South Wales secondary schools by broadly
identifying the critical leadership behaviours practised in schools and their influence
on teacher perceptions of school learning environment and teacher outcomes. Study
2 has clarified the way in which these relations function with reference to three case
study schools, providing evidence linking leadership style behaviours, several school
learning environment components, and selected teacher outcomes. The combination
of quantitative and qualitative findings together also underscores the
multidimensional nature of these relations, and provides rich insight into their
functionality.
Several important conclusions emerge as a result of the quantitative analysis
undertaken in Study 1. Firstly, teachers who participated in the survey indicated their
strong preference for and overall satisfaction with principals who exercised
leadership based on individualised consideration style behaviours rather than on
visionary leadership style behaviours. This finding is contrary to the transformational
and transactional leadership literature, which suggests that the most potent motivator
to performance beyond expectation among followers is through raising followers’
internal commitment to a transcendental vision (Bass & Avolio, 1997). Rather, this
finding demonstrates that schools are human communities that are relationally based,
and that members in these communities respond to individualised attention. The one
303
school learning environment factor where individualised consideration style
behaviour had a major influence was affiliation, which was a measure of the
perception of staff relationships or collegiality. Staff who experienced individualised
consideration leadership behaviours from their principals also experienced higher
perceptions of staff collegiality than staff whose principals practised visionary
leadership behaviours. Paradoxically, visionary leadership behaviour had a
statistically negligible effect on six of the seven school learning environment
constructs examined in this investigation, and no statistically significant effects on
any of the four teacher outcomes under consideration, a finding supportive of the
relationship-based nature of school communities. A third important finding of Study
1 focused on the role of laissez-faire leadership behaviours. In almost every case, a
principal who practised laissez-faire leadership style behaviours had a negative
influence on teachers’ perceptions of school learning environment. This is
particularly worrisome in that principals who fail to show strong leadership
behaviours by avoiding decision-making, being absent or unavailable to staff when
needed, or failing to intervene until an issue becomes a problem, risk staff finding
alternative or substitutionary leadership and losing the collegial unity that comes
from staff having a common vision or purpose.
The qualitative phase of this present study, Study 2, highlighted several
important principles relating to the formation and implementation of a school vision,
and practical methods that school leaders can use to demonstrate genuine
individualised consideration behaviours to their staff. Further, a set of strategies has
emerged that principals can use to directly enhance the school learning environments
of their schools. Firstly, for vision to be effective, it must be reviewed and revisited
by the leader on a regular basis. New staff must be encouraged to engage with the
meaning and importance of the vision for their learning communities, and principals
must be prepared to allow fresh interpretations and applications of the vision to
occur. This encourages inclusiveness, ownership and commitment to the vision.
Several principal leadership behaviours were also identified as being germane to
encouraging individualised consideration. These included a well-publicised open
door policy, such that staff feel they have access to the leader; being seen around the
school—especially at events designed to strengthen community; and recognising and
celebrating success in all its forms, both privately and publicly.
304
The synergistic combination of quantitative and qualitative methods
employed in this study helped elucidate those principal leadership practices
responsible for influencing teacher perceptions of school learning environment and
teacher outcomes within the context of New South Wales’ secondary schools.
Further, this multimethod approach helped identify those mechanisms by which
principal leadership transmits influence through a school community. The results
also identify the potential problems associated with the reliance on one particular
style of leadership behaviour over another, and point to the limited influence that one
leadership style alone can have as far as the management of a school is concerned.
As such, the substantive and methodological implications of this study and the
ramifications for practitioners are important. This study has challenged the assumed
primacy of commitment to vision over individualised consideration behaviours, as a
motivator to teacher performance. Clear lines of influence have been demonstrated
between key leadership style behaviours, selected school learning environment
measures and teacher outcomes, indicating that principals can influence both
environment and outcome variables within their schools. Finally, both visionary and
individualised consideration style leadership behaviours can be taught, giving new
professional development possibilities for principals working in schools.
305
APPENDICES
322
APPENDIX 4.1
PERMISSION TO CONDUCT RESEARCH
323
324
325
APPENDIX 4.2
INSTRUCTIONS REGARDING
INVOLVEMENT IN THIS RESEARCH
4.2.1. Invitation to participate in Research Project
FAX DATE: Number of Pages including cover: TIME: TO:
FROM: Alan M. Barnett. B.A., Dip.Ed., M.Ed. Admin., M.A.C.E.
“The Leadership Style Behaviour of School Principals, School Learning Environments and Aspects of Teacher Job Satisfaction.”
MESSAGE: Dear Dr/Mr/Mrs/M/s Thank you for showing interest in this research project. Please find attached a brief summary description of my proposal. I would be pleased to answer any questions you may have. I seek your approval to conduct this research in your school. I will contact you in a few days, to answer any questions you may have. Again, thank you for your time. Alan M. Barnett Researcher
326
4.2.2. Instructions sent with Staff and Principal Questionnaires Dear Principal, Thank you for allowing your School to participate in this research. Please find enclosed a questionnaire booklet and response booklet for you as the Principal to complete. In order to ensure that a representative sample is obtained, I would request that you nominate a member of your teaching staff to randomly select between 10 and 20 teaching staff to complete the questionnaire booklets enclosed. Once you have nominated a member of the teaching staff, would you please pass on to them the following items that are also enclosed: 1. A note on random sampling and a suggested method for selection of teaching staff
to ensure a representative sample. 2. A list of instructions for administering and returning the response booklets. 3. Multiple copies of the questionnaire booklet and response booklet for teaching
staff to complete and individual envelopes to maintain anonymity. 4. A large (A4 size) pre-paid envelope for staff to place completed response booklets for return to me. NB: All Response Booklets have been coded for school identification purposes only. This will enable individual feedback to schools. Yours sincerely, Alan M. Barnett. University of Western Sydney, Macarthur.
327
4.2.3. Instructions regarding Random Sampling procedures Dear Colleague, Thank you for agreeing to fulfil the important role of randomly selecting your colleagues to participate in this research. The following information and instructions are provided to assist you with this process: 1. A note on random sampling and a suggested method for selection of teaching staff
to ensure a random sample.
Random Sampling Random sampling is the process of selecting a sample in such a way that all individuals in a defined population have an equal and independent opportunity of being selected to participate in the study. It is important to note that a random sample means all members of a population have an equal opportunity for selection based on chance. A suggested method is to:
A. Obtain a list of teaching staff at your school. B. Assign them numbers. C. Place the numbers in a hat. D. Pull out at least 15 numbers that represent at least 15 staff members. E. Ask those staff members to complete the questionnaires.
2. After staff have completed the questionnaires, collect and place completed
questionnaire and response booklets into the envelope provided and I will make arrangements to collect it.
3. I recognise that I am asking a lot from you in terms of helping to maintain your colleagues’ anonymity and confidentially. However, providing this is essential for the study and I appreciate your efforts in helping me. Thank you. Yours sincerely, Alan M. Barnett. University of Western Sydney.
328
4.2.4. Letter to Staff thanking them for their involvement
INTRODUCTION
Thank you for taking the time to fill out these questionnaires. Your help is very much appreciated. The goal of this research project is to examine the relationship between School Leadership, School Learning Environment and some Teacher Outcomes. Your participation is invaluable. My desire is to maintain your anonymity and confidentiality. Therefore, I would ask that you in no way identify yourself on the Response Sheet. Please do not mark the Staff Questionnaire Booklet as I intend to reuse it with other participants and in so doing, save some trees. Alan M. Barnett. Researcher. University of Western Sydney,
329
APPENDIX 4.3
TEACHER QUESTIONNAIRES
Multifactor Leadership Questionnaire
(Form 5X—Short) Instructions: This questionnaire is to describe the leadership style of your School Principal as you perceive them. Please answer all items on the Response Sheet provided. If an item is irrelevant, or if you are unsure or do not know the answer, leave the answer blank. Please answer this questionnaire anonymously. Forty-five descriptive statements are listed below. Judge how frequently each statement fits the person you are describing. Use the following rating scale. Not at all Once in a while Sometimes, if not
always Fairly often Frequently
1 2 3 4 5 PLEASE DO NOT RESPOND ON THIS QUESTIONNAIRE. ANSWER ALL
ITEMS ON THE RESPONSE SHEET PROVIDED, BY MARKING THE NUMBER THAT CORRESPONDS TO THE ANSWER OF YOUR CHOICE.
THE PERSON I AM RATING.... 1. Provides me with assistance in exchange for my efforts. 2. Re-examines critical assumptions to question whether they are appropriate. 3. Fails to interfere until problems become serious. 4. Focuses attention on irregularities, mistakes, exceptions, and deviations from standards. 5. Avoids getting involved when important issues arise. 6. Talks about their most important values and beliefs regarding education.
330
7. Is absent when needed. 8. Seeks differing perspectives when solving problems. 9. Talks optimistically about the future. 10. Instils pride in me for being associated with him/her. 11. Discusses in specific terms who is responsible for achieving performance targets. 12. Waits for things to go wrong before taking action. 13. Talks enthusiastically about what needs to be accomplished. 14. Specifies the importance of having a strong sense of purpose. 15. Spends time teaching and coaching. 16. Makes clear what one can expect to receive when performance goals are achieved. 17. Shows that he/she is a firm believer in, “If it ain’t broke, don’t fix it.” 18. Goes beyond self interest for the good of the group. 19. Treats me as an individual rather than just a member of the group. 20. Demonstrates that problems must become chronic before taking action. 21. Acts in ways that builds my respect. 22. Concentrates his/her full attention on dealing with mistakes, complaints and failures. 23. Considers the moral and ethical consequences of decisions. 24. Keeps track of all mistakes. 25. Displays a sense of power and confidence. 26. Articulates a compelling vision for the future. 27. Directs my attention toward failures to meet standards. 28. Avoids making decisions. 29. Considers me as having different needs, abilities and aspirations from others.
331
30. Gets me to look at problems from many different angles. 31. Helps me to develop my strengths. 32. Suggests new ways of looking at how to complete assigned tasks. 33. Delays responding to urgent questions. 34. Emphasises the importance of having a collective sense of mission. 35. Expresses satisfaction when I meet expectations. 36. Expresses confidence that goals will be achieved. 37. Is effective in meeting my job related needs. 38. Uses methods of leadership that are satisfying. 39. Gets me to do more than I expected to do. 40. Is effective in representing me to higher authority. 41. Works with me in a satisfactory way. 42. Heightens my desire to succeed. 43. Is effective in meeting organisational requirements. 44. Increases my willingness to try harder. 45. Leads a group that is effective.
332
School Level Environment Questionnaire
Instructions: There are 56 items in this questionnaire. They refer to statements about the School in which you work and your working environment. Think about how well the statements describe your School environment. Indicate your answers using the following descriptors:
Strongly Agree
Agree Neither Agree or Disagree
Disagree Strongly Disagree
1 2 3 4 5
PLEASE DO NOT RESPOND ON THIS QUESTIONNAIRE. ANSWER ALL ITEMS ON THE RESPONSE SHEET PROVIDED.
1. There are many disruptive, difficult students in the School. 2. I seldom receive encouragement from colleagues. 3. Teachers frequently discuss teaching methods and strategies with each other. 4. I am often supervised to ensure that I follow directions correctly. 5. Decisions about running the School are often made by the Principal or a small
group of teachers. 6. It is very difficult to change anything in this School. 7. The School or Department library includes an adequate selection of books and periodicals. 8. There is constant pressure to keep working. 9. Most students are helpful and co-operative to teachers. 10. I feel accepted by other teachers. 11. Teachers avoid talking with each other about teaching and learning.
333
12. I am not expected to conform to a particular teaching style. 13. I have to refer even small matters to a senior member of staff for a final answer. 14. Teachers are encouraged to be innovative in this School. 15. The supply of equipment and resources is inadequate. 16. Teachers have to work long hours to complete their work. 17. Most students are pleasant and friendly to teachers. 18. I am ignored by other teachers. 19. Professional matters are rarely discussed at staff meetings. 20. It is considered very important that I closely follow syllabuses and lesson plans. 21. Actions can usually be taken without gaining the approval of the subject department head or a senior member of staff. 22. There is a great deal of resistance to proposals for curriculum change. 23. Video equipment, tapes and films are readily available and accessible. 24. Teachers don’t have to work very hard in this School. 25. There are many noisy, badly behaved students. 26. I feel I could rely on my colleagues for assistance if I should need it. 27. Many teachers attend inservice and other professional development courses. 28. There are a few rules and regulations that I am expected to follow. 29. Teachers are frequently asked to participate in decisions concerning administrative policies and procedures. 30. Most teachers like the idea of change. 31. Adequate duplicating facilities and services are available to teachers. 32. There is not time for teachers to relax. 33. Students get along well with teachers. 34. My colleagues seldom take notice of my professional views and opinions. 35. Teachers show little interest in what is happening in other schools .
334
36. I am allowed to do almost as I please in the classroom. 37. I am encouraged to make decisions without reference to a senior member of staff. 38. New courses or curriculum materials are seldom implemented in the School. 39. Tape recorders and cassettes are seldom available when needed. 40. You can take it easy and still get the work done. 41. Most students are well mannered and respectful to the School staff. 42. I feel that I have many friends among my colleagues at this School. 43. Teachers are keen to learn from their colleagues. 44. My classes are expected to use prescribed textbooks and prescribed resource material. 45. I must ask my subject department head or senior member of staff before I do
most things. 46. There is much experimentation with different teaching approaches. 46. Facilities are inadequate for catering for a variety of classroom activities and learning groups of different sizes. 48. Seldom are there deadlines to be met. 49. Very strict discipline is needed to control many of the students. 50. I often feel lonely and left out of things in the staffroom. 51. Teachers show considerable interest in the professional activities of their colleagues. 52. I am expected to maintain very strict control in the classroom. 53. I have very little say in the running of the School. 54. New and different ideas are always being tried out in this School. 55. Video’s, overhead transparencies and access to photocopying are usually
available when needed. 56. It is hard to keep up with your workload.
335
Teacher Outcomes Questionnaire Instructions: There are 6 items in this questionnaire. They refer to statements about how you feel about the work you do in your current school. Think about how well the statements describe your School environment. Please ignore the item numbers at the end of each statement. Indicate your answers using the following descriptors: Strongly Disagree
Disagree Neither Disagree or Agree
Agree Strongly Agree
1 2 3 4 5 1. In this school, some teachers have more influence than other teachers. (Item 4) 2. In this school, the administration shows favouritism to some teachers. (Item 11) 3. I can deal with almost any learning problem. (Item 21) 4. I am certain I am making a difference in the lives of my students. (Item 28) 5. Power and influence count a lot around this school. (Item 38) 6. I am good at helping all the students in my class make significant improvement.
(Item 39)
336
Demographic Information
Instructions: Mark the number corresponding to the answer of your choice on the Answer Sheet provided. 1. What is your current position? 1 Part-time teacher 4 Deputy Principal 2 Full-time teacher 5 Leading Teacher 3 Head Teacher 6 Other (specify) 2. How many years have you been in the teaching profession? 1 1 to 2 years 4 11 to 20 years 2 3 to 5 years 5 20+ years 3 6 to 10 years 3. How many years have you been at your present school? 1 1 to 2 years 4 11 to 20 years 2 3 to 5 years 5 20+ years 3 6 to 10 years 4. How many students are in your school? 1 Less than 200 4 601—800 2 201— 400 5 801—1000 3 401— 600 6 1001—1200 7 More than 1201 5. Your age? 1 Less than 30 years 3 40 to 49 years 2 30 to 39 years 4 50 to 59 years 5 60 + years 6. Your gender? 1 Male 2 Female 7. The School Principal I am describing is a: 1 Male 2 Female 8. How long have you worked with the School Principal you are describing? 1 3 months or less 4 between 1 and 2 years 2 between three and six months 5 over two years 3 between six and twelve months
337
MULTIFACTOR LEADERSHIP QUESTIONNAIRE
TEACHER RESPONSE SHEET Not at all Once in a while Sometimes, if not
always Fairly often Frequently
1 2 3 4 5
1 1 2 3 4 5 24 1 2 3 4 5
2 1 2 3 4 5 25 1 2 3 4 5
3 1 2 3 4 5 26 1 2 3 4 5
4 1 2 3 4 5 27 1 2 3 4 5
5 1 2 3 4 5 28 1 2 3 4 5
6 1 2 3 4 5 29 1 2 3 4 5
7 1 2 3 4 5 30 1 2 3 4 5
8 1 2 3 4 5 31 1 2 3 4 5
9 1 2 3 4 5 32 1 2 3 4 5
10 1 2 3 4 5 33 1 2 3 4 5
11 1 2 3 4 5 34 1 2 3 4 5
12 1 2 3 4 5 35 1 2 3 4 5
13 1 2 3 4 5 36 1 2 3 4 5
14 1 2 3 4 5 37 1 2 3 4 5
15 1 2 3 4 5 38 1 2 3 4 5
16 1 2 3 4 5 39 1 2 3 4 5
17 1 2 3 4 5 40 1 2 3 4 5
18 1 2 3 4 5 41 1 2 3 4 5
19 1 2 3 4 5 42 1 2 3 4 5
20 1 2 3 4 5 43 1 2 3 4 5
21 1 2 3 4 5 44 1 2 3 4 5
22 1 2 3 4 5 45 1 2 3 4 5
23 1 2 3 4 5
338
SCHOOL LEVEL ENVIRO NMENT QUESTIONNAIRE
TEACHER RESPONSE SHEET
Strongly Agree
Agree Neither Agree or Disagree
Disagree Strongly Disagree
1 2 3 4 5
1 1 2 3 4 5 29 1 2 3 4 5
2 1 2 3 4 5 30 1 2 3 4 5
3 1 2 3 4 5 31 1 2 3 4 5
4 1 2 3 4 5 32 1 2 3 4 5
5 1 2 3 4 5 33 1 2 3 4 5
6 1 2 3 4 5 34 1 2 3 4 5
7 1 2 3 4 5 35 1 2 3 4 5
8 1 2 3 4 5 36 1 2 3 4 5
9 1 2 3 4 5 37 1 2 3 4 5
10 1 2 3 4 5 38 1 2 3 4 5
11 1 2 3 4 5 39 1 2 3 4 5
12 1 2 3 4 5 40 1 2 3 4 5
13 1 2 3 4 5 41 1 2 3 4 5
14 1 2 3 4 5 42 1 2 3 4 5
15 1 2 3 4 5 43 1 2 3 4 5
16 1 2 3 4 5 44 1 2 3 4 5
17 1 2 3 4 5 45 1 2 3 4 5
18 1 2 3 4 5 46 1 2 3 4 5
19 1 2 3 4 5 47 1 2 3 4 5
20 1 2 3 4 5 48 1 2 3 4 5
21 1 2 3 4 5 49 1 2 3 4 5
22 1 2 3 4 5 50 1 2 3 4 5
23 1 2 3 4 5 51 1 2 3 4 5
24 1 2 3 4 5 52 1 2 3 4 5
25 1 2 3 4 5 53 1 2 3 4 5
26 1 2 3 4 5 54 1 2 3 4 5
27 1 2 3 4 5 55 1 2 3 4 5
28 1 2 3 4 5 56 1 2 3 4 5
339
TEACHER OUTCOMES QUESTIONNAIRE
TEACHER RESPONSE SHEET
Strongly Disagree
Disagree Neither Disagree or Agree
Agree Strongly Agree
1 2 3 4 5
1 1 2 3 4 5
2 1 2 3 4 5
3 1 2 3 4 5
4 1 2 3 4 5
5 1 2 3 4 5
6 1 2 3 4 5
340
DEMOGRAPHIC INFORMATION
TEACHER RESPONSE SHEET
1 1 2 3 4 5 6
2 1 2 3 4 5
3 1 2 3 4 5
4 1 2 3 4 5 6 7
5 1 2 3 4 5
6 1 2
7 1 2
8 1 2 3 4 5
THANK YOU FOR PARTICIPATING
IN THIS RESEARCH
341
APPENDIX 4.4
PRINCIPAL QUESTIONNAIRE
Multifactor Leadership Questionnaire
(Form 5X—Short) Instructions: This questionnaire describes your leadership style as a School Principal. Please answer all items on the Response Sheet provided. If an item is irre levant, or if you are unsure or do not know the answer, leave the answer blank. Please answer this questionnaire anonymously. Forty-five descriptive statements are listed below. Judge how frequently each statement fits you. Not at all Once in a while Sometimes, if not
always Fairly often Frequently
1 2 3 4 5 PLEASE DO NOT RESPOND ON THIS QUESTIONNAIRE. ANSWER ALL
ITEMS ON THE RESPONSE SHEET PROVIDED, BY MARKING THE NUMBER THAT CORRESPONDS TO THE ANSWER OF YOUR CHOICE.
1. I provide others with assistance in exchange for their efforts. 2. I re-examine critical assumptions to question whether they are appropriate. 3. I fail to interfere until problems become serious. 4. I focus attention on irregularities, mistakes, exceptions, and deviations from standards. 5. I avoid getting involved when important issues arise. 6. I talk about my most important values and beliefs. 7. I am absent when needed. 8. I seek differing perspectives when solving problems. 9. I talk optimistically about the future.
342
10. I instill pride in others for being associated with me. 11. I discuss in specific terms who is responsible for achieving performance targets. 12. I wait for things to go wrong before taking action. 13. I talk enthusiastically about what needs to be accomplished. 14. I specify the importance of having a strong sense of purpose. 15. I spend time teaching and coaching. 16. I make clear what one can expect to receive when performance goals are achieved. 17. I show that I am a firm believer in, “If it ain’t broke, don’t fix it.” 18. I go beyond self interest for the good of the group. 19. I treat others as individuals rather than just as a member of a group. 20. I demonstrate that problems must become chronic before I take action. 21. I act in ways that build other’s respect for me. 22. I concentrate my full attention on dealing with mistakes, complaints and failures. 23. I consider the moral and ethical consequences of decisions. 24. I keep track of all mistakes. 25. I display a sense of power and confidence. 26. I articulate a compelling vision for the future. 27. I direct my attention toward failures to meet standards. 28. I avoid making decisions. 29. I consider an individual as having different needs, abilities and aspirations from others. 30. I get others to look at problems from many different angles. 31. I help others to develop their strengths. 32. I suggest new ways of looking at how to complete assigned tasks. 33. I delay responding to urgent questions.
343
34. I emphasise the importance of having a collective sense of mission. 35. I express satisfaction when others meet expectations. 36. I express confidence that goals will be achieved. 37. I am effective in meeting others’ job related needs. 38. I use methods of leadership that are satisfying. 39. I get others to do more than they expected to do. 40. I am effective in representing others to higher authority. 41. I work with others in a satisfactory way. 42. I heighten others’ desire to succeed. 43. I am effective in meeting organisational requirements. 44. I increase others’ willingness to try harder. 45. I lead a group that is effective.
344
Demographic Information (Principals) Instructions : Mark the number corresponding to the answer of your choice on the Answer Sheet provided. 1. How many years have you been in the teaching profession? 1 1 to 2 years 4 11 to 20 years 2 3 to 5 years 5 20+ years 3 6 to 10 years 2. How many years have you been at your present school? 1 1 to 2 years 4 11 to 20 years 2 3 to 5 years 5 20+ years 3 6 to 10 years 3. How many students are in your school? 1 Less than 200 4 601—800 2 201— 400 5 801—1000 3 401— 600 6 1001—1200 7 More than 1201 4. Your age? 1 Less than 30 years 3 40 to 49 years 2 30 to 39 years 4 50 to 59 years 5 60 + years 5. Your gender? 1 Male 2 Female 6. How many years have you been a School Principal? 1 1 to 4 years 5 10 to 12 years 2 4 to 6 years 6 12 to 14 years 3 6 to 8 years 7 14 to 16 years
4 8 to 10 years 8 more than 16 years 7. What is your current level of education? 1 Undergraduate Degree 5 Ed.D. 2 Graduate Certificate 6 Ph.D. 3 Graduate Diploma 7 Other (please specify)
4. Masters Degree 8. I am currently involved in further study at the level of: 1 Undergraduate Degree 5 Ed.D. 2 Graduate Certificate 6 Ph.D. 3 Graduate Diploma 7 Other (please specify)
4. Masters Degree
345
MULTIFACTOR LEADERSHIP QUESTIONNAIRE
PRINCIPAL’S RESPONSE SHEET Not at all Once in a while Sometimes, if not
always Fairly often Frequently
1 2 3 4 5
1 1 2 3 4 5 24 1 2 3 4 5
2 1 2 3 4 5 25 1 2 3 4 5
3 1 2 3 4 5 26 1 2 3 4 5
4 1 2 3 4 5 27 1 2 3 4 5
5 1 2 3 4 5 28 1 2 3 4 5
6 1 2 3 4 5 29 1 2 3 4 5
7 1 2 3 4 5 30 1 2 3 4 5
8 1 2 3 4 5 31 1 2 3 4 5
9 1 2 3 4 5 32 1 2 3 4 5
10 1 2 3 4 5 33 1 2 3 4 5
11 1 2 3 4 5 34 1 2 3 4 5
12 1 2 3 4 5 35 1 2 3 4 5
13 1 2 3 4 5 36 1 2 3 4 5
14 1 2 3 4 5 37 1 2 3 4 5
15 1 2 3 4 5 38 1 2 3 4 5
16 1 2 3 4 5 39 1 2 3 4 5
17 1 2 3 4 5 40 1 2 3 4 5
18 1 2 3 4 5 41 1 2 3 4 5
19 1 2 3 4 5 42 1 2 3 4 5
20 1 2 3 4 5 43 1 2 3 4 5
21 1 2 3 4 5 44 1 2 3 4 5
22 1 2 3 4 5 45 1 2 3 4 5
23 1 2 3 4 5
346
DEMOGRAPHIC INFORMATION
PRINCIPAL’S RESPONSE SHEET
1 1 2 3 4 5
2 1 2 3 4 5
3 1 2 3 4 5 6 7
4 1 2 3 4 5
5 1 2
6 1 2 3 4 5 6 7 8
7 1 2 3 4 5 6 7
8 1 2 3 4 5 6 7
We would welcome any comments you would like to make: __________________________________________________________________________________
__________________________________________________________________________________
__________________________________________________________________________________
__________________________________________________________________________________
__________________________________________________________________________________
__________________________________________________________________________________
THANK YOU FOR PARTICIPATING
IN THIS RESEARCH
347
APPENDIX 4.5
INTERVIEW PROTOCOLS:
TEACHERS AND PRINCIPAL
Interview Protocols—School Leadership/School Learning Environment
TEACHER QUESTIONS
Aspects of Leadership
(Bracketed – shows relation of question to either leadership, school learning environment or teacher outcome measure)
1. Does your school have a vision? What are the main elements of that vision?
(Transformation Leadership – Idealised Influence Attributed)
Prompt: What is your school’s vision?
2. Did you have input in the development of the school’s vision? If so, how?
(Transformational Leadership – Idealised Influence Behaviour )
Prompt: Does the staff own the vision?
Prompt: Do you own the vision?
Prompt: Do the other stakeholders share the vision?
3. What strategies are employed in the implementation of the vision?
(Transformational Leadership – Idealised Influence Behaviour)
Prompt: What does the Principal do to ensure that the school’s vision is
achieved?
4. How central is the school’s vision in the life of the school?
(Transformational Leadership – Idealised Influence Attributed)
Prompt: How does the Principal defend the vision?
Prompt: How is opposition to the vision dealt with in the school?
5. Do you feel the leader (Principal) knows you?
(Transactional Leadership – Contingent Reward)
Prompt: What strategies does the Principal use to get to know the staff?
6. If you wanted to talk to the Principal about something personal or educational,
how would you go about doing this?
(Transactional Leadership – Contingent Reward)
Prompt: Do you have access to the Principal?
Prompt: Is the Principal approachable? How does he/she demonstrate this?
348
Aspects of Teacher Outcomes
7. What strategies does the Principal use to encourage excellence?
(Transformational Leadership – Idealised Influence Behaviour)
Prompt: How does the Principal encourage you to work hard?
8. How does the Principal let you know you have done a good job?
(Teacher Satisfaction – Global Satisfaction)
Prompt: How does the Principal make you feel valued within the School?
9. What are the signs of excellence that the Principal looks for in his/her staff?
(Teacher Satisfaction – Effectiveness)
10. What makes working at this school satisfying?
(Teacher Satisfaction – Global Satisfaction)
Prompt: Is this a good school to teach in? Why?
Aspects of School Learning Environment
11. What attitudes and values are important in this school as far as the learning
environment is concerned?
(School Learning Environment – all variables)
Prompt: How would you describe the learning environment in the school? What
things (factors) are important?
Prompt: What are some of the elements that make this a good school to teach and
learn in?
12. Does the staff in this school want to be here? Do the students of this school want
to be here?
(School Learning Environment – Student Supportiveness and Affiliation)
Prompt: Do staff feel supported in this school? How has this been demonstrated?
Prompt: Do students feel supported in this school? How has this been
demonstrated?
13. What things have been done to create a positive learning environment?
(School Learning Environment – Formalisation and Centralisation)
Prompt: How has the Principal affected the learning environment? How have the
students affected the learning environment? How has staff affected the learning
environment?
349
14. What effects do changes or innovations have of the learning environment?
(School Learning Environment – Innovation)
Prompt: How are new ideas/methods/resources received in this school by staff?
Students?
15. In what ways are you involved in the decision making processes in this school?
(School Learning Environment – Formalisation and Centralisation)
Prompt: At what level within the school are decisions made? (i.e., homework
policy, extra’s policy, etc) Do you think you could change the school’s policy in
these areas (homework, extras, etc)? How would you go about changing school
policy?
16. How would you go about getting more teaching resources or school resources in
the school?
(School Learning Environment – Formalisation, Centralisation and Resource
Adequacy)
Prompt: Teaching Resources – OHP’s, videos, textbooks, etc.
Prompt: School Resources – shade cover, fencing, playground equipment, etc.
350
Interview Protocols—School Leadership/School Learning Environment
PRINCIPAL QUESTIONS
Aspects of Leadership
1. Does your school have a vision? What are the main elements of that vision?
(Transformational Leadership - Idealised Influence Attributed)
Prompt: What is your school’s vision?
2. Who was involved in the development of the school’s vision? If so, how?
((Transformational Leadership - Idealised Influence Behaviour)
Prompt: Does the staff own the vision?
Prompt: Do the students own the vision?
Prompt: Do the other stakeholders share the vision?
3. What strategies are employed in the implementation of the vision?
(Transformational Leadership – Idealised Influence Behaviour)
Prompt: What do you do to ensure that the school’s vision is achieved?
4. How central is the school’s vision in the life of the school? (TF)
(Transformational Leadership – Idealised Influence Attributed)
Prompt: How do you defend the vision?
Prompt: How is opposition to the vision dealt with in the school?
5. Do you feel you know your staff?
(Transactional Leadership – Contingent Reward)
Prompt: What strategies do you use to get to know the staff?
6. If the staff wanted to talk to with you about something personal or educational,
how would they go about doing this?
(Transactional Leadership – Contingent Reward)
Prompt: Does the staff have access to the Principal?
Prompt: Are you approachable? How do you demonstrate this?
Aspects of Teacher Outcomes
7. What strategies do you use to encourage excellence?
(Transformational Leadership – Idealised Influence Attributed)
Prompt: How do you encourage you to work hard?
351
8. How do you let the staff know they have done a good job?
(Teacher Satisfaction – Global Satisfaction)
Prompt: How do you make the staff feel valued within the School?
9. What are the signs of excellence that you look for in his/her your staff?
(Teacher Satisfaction – Effectiveness)
10. What makes working at this school satisfying?
(Teacher Satisfaction – Global Satisfaction)
Prompt: Is this a good school to teach in? Why?
Aspects of School Learning Environment
11. What attitudes and values are important in this school as far as the learning
environment is concerned?
(School Learning Environment – all variables)
Prompt: How would you describe the learning environment in the school? What
things (factors) are important?
Prompt: What are some of the elements that make this a good school to teach and
learn in?
Prompt: What practices do you use to make this a better learning place for students?
Working place for teachers?
12. Does the staff in this school want to be here? Do the students of this school want
to be here?
(School Learning Environment – Student Supportiveness and Affiliation)
Prompt: Does staff feel supported in this school? How has this been demonstrated?
Prompt: Do students feel supported in this school? How has this been demonstrated?
13. What things have been done to create a positive learning environment?
(School Learning Environment – Formalisation and Centralisation)
Prompt: How has the Principal affected the learning environment? How have the
students affected the learning environment? How has staff affected the learning
environment?
14. What effects do changes or innovations have of the learning environment?
(School Learning Environment – Innovation)
Prompt: How are new ideas/methods/resources received in this school by staff?
Students?
352
15. In what ways are you involved in the decision making processes in this school?
(School Learning Environment – Formalisation, Centralisation)
Prompt: At what level within the school are decisions made? (ie, homework policy,
extra’s policy, etc) Do you think you could change the school’s policy in these areas
(homework, extras, etc)? How would you go about changing school policy?
16. How would you go about getting more teaching resources or school resources in
the school?
(School Learning Environment – Formalisation, Centralisation and Resource
Adequacy)
Prompt: Teaching Resources – OHP’s, videos, textbooks, etc.
Prompt: School Resources – shade cover, fencing, playground equipment, etc.
17. How do you encourage staff involvement in the decision making process?
(School Learning Environment – Formalisation, Centralisation)
353
APPENDIX 4.6
INFORMED CONSENT
“The Leadership Style Behaviour of School Principals, School Learning Environments and Aspects of Teacher Job Satisfaction.”
Informed Consent
Thank you for taking the time to complete this survey. It aims to identify those leadership behaviours which enhance learning and working within your school. You have been approached to participate in this research because of your years of service in this school and therefore your intimate knowledge of this school’s “personality”. I would ask that you complete the questionnaires attached, not spending too much time on any one question, but rather moving through them all reasonably quickly. It should take no more than 30 minutes to complete the instruments. After completing the questionnaire, please seal your responses in the envelope provided, and return to the survey co-ordinator in your School. Please respond as openly and truthfully as you can, taking special care not to write your name anywhere on the data response sheet or the envelope provided. Your identity will remain completely anonymous throughout the course of the research project. Your responses will be added to other responses collected from within your School. Further, both your Principal and your School will not be directly identified throughout the course of this project. Data collected from this project will only be used for the purposes outlined above, maintaining both your confidentiality and privacy. Your participation is voluntary and you may withdraw at any time without prejudice. This type of study involves no “treatment” or “control” group, and therefore no placebos are used in its design. Again, your participation is greatly appreciated. Your responses, along with the other responses collected, will provide a picture of those leadership practices within schools that make them positive environments in which to work and learn. I would be pleased to return to your School at some future time to share the results of your participation. Yours sincerely, Alan Barnett. B.A., Dip.Ed., M.Ed.Admin., M.A.C.E. Researcher. This Research project is being undertaken with the consent of the University of Western Sydney, Macarthur, Ethics Review Committee (Human Subjects).
354
APPENDIX 5.1
DEVELOPMENT OF LEADERS HIP, SCHOOL LEARNING
ENVIRONMENT AND TEACHER OUTCOMES MODELS
Introduction
A full discussion of the influence of differential leadership styles on elements
of school learning environment and teacher outcomes requires firstly that the factor
structure for each of the leadership, school learning environment and teacher
outcomes models be established. This appendix outlines the development of model
structures for each of the areas of leadership, school learning environment and
teacher outcomes.
Transformational and Transactional Leadership
Exploratory factor analysis
The Multifactor Leadership Questionnaire 5X (Short) is a 36 item instrument
designed to examine five transformational leadership constructs, four transactional
leadership constructs and one non leadership construct (Bass et al., 1997; Table
5.1.1). An initial analysis of the thirty six item questionnaire produced six factors
with eigenvalues greater than 1, and variation and cumulative percentages as
indicated in Table 5.1.2 below. A further analysis indicated two factors (factors 5 and
6) to be uninterpretable. A final analysis, using principal axis factoring and oblimin
rotation produced four identifiable factors, with eigenvalues and variance shown in
Table 5.1.3. The scree plot is shown as Figure 5.1.1 below.
355
5.1.1
Bass et al.’s (1997) Nine-factor model, Item numbers and Cronbach’s Alpha
Leadership Style Construct Item Numbers Cronbach’s
Alpha Transformational Idealised Influence—
Behaviour 6, 14, 23, 34 .71
Transformational Idealised Influence—Attributed
10, 18, 21, 25 .56
Transformational Inspirational Motivation 9, 13, 26, 36 .82 Transformational Intellectual Stimulation 2, 8, 30, 32 .79 Transformational Individualised Consideration 15, 19, 29, 31 .80 Transactional Contingent Reward 1, 11, 16, 35 .77 Transactional Management by
Exception—Active 4, 22, 24, 27 .60
Transactional Management by Exception—Passive
3, 12, 17, 20 .67
Non-Transactional Laissez-Faire 5, 7, 28, 33 .76 Table 5.1.2
MLQ-5X Eigenvalues, Variation and Cumulative Percentage
Factor Number
Eigenvalues Percent of Variation
Cumulative Percentage
1 12.84 35.7 35.7 2 2.35 6.5 42.2 3 2.15 6.0 48.2 4 1.66 4.6 52.8 5 1.19 3.3 56.1 6 1.12 3.1 59.2 7 .94 2.6 61.8 8 .86 2.4 64.2 9 .85 2.4 66.6
Four factors were identified from this 36-item data set. Beyond these four
factors, eigenvalues were less than one, and interpretability became difficult.
Table 5.1.3
MLQ-5X Eigenvalues, Variation and Cumulative Percentage
Factor Number
Eigenvalues Percent of Variation
Cumulative Percentage
1 12.41 34.47 34.47 2 1.85 5.14 39.61 3 1.60 4.45 44.05 4 1.19 3.31 47.36
356
Scree Plot
Factor Number
3533312927252321191715131197531
Eig
env
alu
e14
12
10
8
6
4
2
0
Figure 5.1.1. MLQ-5X leadership factors scree plot.
The first unrotated factor, explained 34.47% of the variation in the data set
(Table 5.1.3). After rotation, the first rotated factor was identified as “individualised
consideration” (IC), and contained items that emphasise the leader’s interest in the
follower as an individual. Items such as “treats me as an individual rather than just
a member of the group” (Item 19) and “considers me as having different needs,
abilities and aspirations from others” (Item 29) reflect the leader’s concern for
followers as individuals, not just a members of a larger group. This factor contains
items from Bass et al.’s (1997) original individualised consideration construct. This
construct lies at the lower end of those factors Bass et al. (1997) called
transformational leadership style behaviours, and is placed on a continuum between
the higher order transformational styles of inspirational motivation and intellectual
simulation, and the transactional style of contingent reward. Items such as “re-
examines critical assumptions to question whether they are appropriate” (Item 2)
and “gets me to look at problems from many different angles” (Item 30) are included
in this factor.
Table 5.1.4 presents the pattern matrix resulting from an analysis of the 36
item, nine factor MLQ-5X (Short). Only the first six factors are shown with loadings
greater than .30. Table 5.1.5 shows the structure matrix for this same data set, with
loadings and cross loadings greater than .30 shown.
357
Table 5.1.4 MLQ-5X Pattern Matrix
Factors
1 2 3 4 5 6
M31 .89 M30 .83 M32 .74 M29 .68 M1 .56 M10 .56 M19 .55 .32 M21 .49 .31 M15 .44 M18 .42 M35 .41 M2 .35 M16 .31 M28 .75 M12 .72 M33 .70 M20 .65 M7 .60 M3 .57 M5 .54 M25 .40 M24 .69 M27 .50 M22 .49 M4 .41 M14 -.74 M9 -.69 M13 -.68 M34 -.54 M26 -.52 M6 -.51 M36 -.48 M8 -.31 M11 -.30 M23 .48 M17 -.33 Extraction Method: Principal Axis Analysis Rotation Method: Oblimin Rotation converged in 35 iterations
358
Table 5.1.5
MLQ-5X Structure Matrix
Factors
1 2 3 4 5 6
M31 .88 -.47 -.41 M30 .79 -.36 -.45 M10 .78 -.51 -.59 .32 M21 .75 -.61 -.57 .54 M1 .75 -.48 -.52 .35 M32 .74 -.39 -.47 M19 .70 -.41 -.39 .51 M18 .68 -.50 -.55 .43 M29 .67 -.30 -.34 M35 .66 -.48 -.49 .45 M2 .61 -.45 -.58 .35 M16 .59 -.49 -.56 -.36 M15 .54 -.37 -.36 M12 -.41 .78 .35 M20 -.40 .72 .31 -.33 M28 -.31 .70 .31 M33 -.35 .69 M7 -.40 .64 .32 M5 -.36 .62 .39 M3 -.35 .60 M25 .48 -.31 .45 M24 .70 M27 .51 -.36 M4 .48 -.30 M22 .46 M14 .47 -.42 -.79 M13 .48 -.39 -.73 M9 .41 -.36 -.71 .30 M34 .50 -.43 -.67 .38 M26 .51 -.47 -.67 M36 .55 -.46 -.67 .37 M8 .51 -.36 -.51 .38 M11 .40 -.42 -.49 M6 -.46 M23 .48 -.40 -.51 .58 M17 -.35 Extraction Method: Principal Axis Analysis Rotation Method: Oblimin
359
Table 5.1.6 indicates the correlation statistics of the six factors extracted from
the data set that had eigenvalues greater than 1. While the correlations as presented
are acceptable, an analysis of some factors’ cross loading and Cronbach alpha values
suggested that a rationalisation of the factor structure was required.
Table 5.1.6
MLQ-5X Factor Correlation Matrix
Factors
1 2 3 4 5 6
1 1.00 2 -.50 1.00 3 -.09 -.12 1.00 4 -.54 .44 -.07 1.00 5 .29 -.25 -.17 -.21 1.00 6 -.14 -.02 -.17 .13 .03 1.00
Extraction Method: Principal Axis Factoring Rotation Method: Oblimin
Table 5.1.7
Leadership Factors, Loadings and Cronbach’s Alpha
Item Description Loading Alpha Factor 1—Individualised Consideration (IC) .94 31. Helps me to develop my strengths. 0.88 30. Gets me to look at problems from many different angles. 0.79 10. Instills pride in me for being associated with him/her. 0.78 21. Acts in ways that builds my respect. 0.75 1. Provides me with assistance in exchange for my efforts. 0.75 32. Suggests new ways of looking at how to complete assigned tasks.
0.75
19. Treats me as an individual rather than just a member of the group.
0.70
18. Goes beyond self-interest for the good of the group. 0.68 29. Considers me as having different needs, abilities and aspirations from others.
0.67
35. Expresses satisfaction when I meet expectations. 0.66 2. Re-examines critical assumptions to question whether they are appropriate.
0.61
16. Makes clear what one can expect to receive when performance goals are achieved.
0.59
15. Spends time teaching and coaching. 0.54
360
Factor 2— Laissez-Faire Leadership (LF) .86 12. Waits for things to go wrong before taking action. 0.77 20. Demonstrates that problems must become chronic before taking action.
0.73
28. Avoids making decisions. 0.70 33. Delays responding to urgent questions. 0.69 7. Is absent when needed. 0.64 5. Avoids getting involved when important issues arise. 0.62 3. Fails to interfere until problems become serious. 0.60 25. Displays a sense of power and confidence. 0.48 Factor 3—Management by Exception—Active (MBEA) .60
24. Keeps track of all mistakes. 0.70 27. Directs my attention toward failures to meet standards. 0.51 4. Focuses attention on irregularities, mistakes, exceptions, and deviations from standards.
0.48
22. Concentrates his/her full attention on dealing with mistakes, complaints and failures.
0.46
Factor 4—Vision (VI) .86 14. Specifies the importance of having a strong sense of purpose. -0.79 13. Talks enthusiastically about what needs to be accomplished. -0.73 9. Talks optimistically about the future. -0.71 34. Emphasises the importance of having a collective sense of mission.
-0.67
26. Articulates a compelling vision for the future. -0.67 36. Expresses confidence that goals will be achieved. -0.67 8. Seeks differing perspectives when solving problems. -0.51 11. Discusses in specific terms who is responsible for achieving performance targets.
-0.49
6. Talks about their most important values and beliefs regarding education.
-0.46
The second unrotated factor, explained 5.14% of the variance (see Table
5.1.3). After rotation, the second factor was identified to be laissez-faire leadership.
It contains items that demonstrate an approach to leadership that emphasises reactive
decision making rather than a proactive style, and is characteristic of leaders who are
often perceived by followers to be absent when needed. Items like “avoids making
decisions” (Item 28) and “is absent when needed” (Item 7) are illustrative, and have
been taken from Bass et al.’s (1997) original laissez-faire or non-leadership style
behaviour.
361
Other items in the laissez-faire leadership (LF) factor include “demonstrates
that problems must become chronic before taking action” (Item 20) and “waits for
things to go wrong before taking action” (Item 12). These items come from Bass et
al.’s (1997) management by exception (passive) construct, and are at the lowest end
of the transformational—transactional leadership continuum.
One anomaly noted is that Item 25 “displays a sense of power and
confidence”, which comes from Bass et al.’s (1997) idealised attributes construct,
has been factored into this laissez-faire Leadership (LF) construct. An explanation
seems difficult, but may well lie in the cultural differences that exist between
Australia and the United States of America, where the instrument was developed. In
Australia, “tall poppies” are cut down in a mostly egalitarian society, whereas in the
United States, they are lauded. Australian teachers may well consider confidence and
power as working against the ability to motivate to higher levels of commitment for
the sake of the group.
The third unrotated factor accounted for 4.45% of the variance (see Table
5.1.3). After rotation, this factor was identified as made up from Bass et al.’s (1997)
management by exception (active; MBEA) construct. Its scales include items that
emphasise the leader’s management of followers by correcting mistakes and meeting
standards. Items include “keeps track of all mistakes” (Item 24) and “directs my
attention toward failures to meet standards” (Item 27).
The final unrotated factor accounted for 3.31% of the variance explained
(Table 5.1.3). After rotation, this factor was identified as consisting of items taken
from Bass et al.’s (1997) original idealised attributes, idealised behaviours and
inspirational motivation constructs. These constructs highlight the leader’s
“charismatic” behaviours, as well as emphasising a preferred future state. Items such
as “articulates a compelling vision for the future” (Item 26), “talks optimistically
about the future” (Item 9) and “specifies the importance of having a strong sense of
purpose” (Item 14) are illustrative. This factor is at the highest point of the
transformational—transactional leadership continuum, where the use of these
leadership behaviours helps lift follower motivation and performance to levels
beyond that which is expected.
362
The exploratory factor analysis produced a pattern matrix (Table 5.1.4) and a
structure matrix (Table 5.1.5) in which cross loadings were identified. Several items
displayed cross loadings of greater than 0.4, which indicate that these items may be
composite in nature, and theoretically associated with more than one factor.
Alternatively, high item cross loadings may indicate that the exploratory factor
analysis has failed to discriminate between factors.
In Factor 1, individualised consideration, Item 35 “expresses satisfaction
when I meet expectations” displayed a high cross loading co-efficient on Factor
2 laissez-faire (-.48) and Factor 4 vision (-0.49), and only loaded to .66 on Factor 1.
This item raises issues of follower satisfaction, and therefore may be applicable at all
levels.
The high cross loading values of Item 18, “goes beyond self-interest for the
good of the group” and Item 16, “makes clear what one can expect to receive when
performance goals are achieved” likewise fail to load clearly on any one factor. This
may be due in part to difficulty in interpretation of the question or to poor item
construction. Item 2, “re-examines critical assumptions to question whether they are
appropriate” and Item 8, “seeks differing perspectives when solving problems” also
display poor discriminate ability.
In Factor 4, vision, Item 34 “emphasises the importance of having a
collective sense of mission” loaded on Factor 1 (IC; 0.50) and Factor 2 (LF; -0.43) as
well as on Factor 4 (VI; -0.67). The same problem also occurred in Item 26,
“articulates a compelling vision for the future” where the item loads on Factor 1 (IC;
0.51), Factor 2 (LF; -.048) and Factor 4 (VI; -0.67).
Item 36, “expresses confidence that goals will be achieved” and Item 11,
“discusses in specific terms who is responsible for achieving performance targets”
also demonstrated cross loading problems. Again, failure to interpret the questions or
poor item construction may be responsible for these results.
363
Model Refining—Leadership
A second exploratory factor analysis was undertaken to determine if the
factors extracted in the first analysis could be further reduced and interpretability
increased, particularly in light of the number of items with high cross loadings.
Two items in the first factor, individualised consideration, cross loaded
highly with Factor 2 (laissez-faire) and Factor 4 (vision). These two Items were 35
and 2 respectively. Further, reliability analysis indicated that these two items could
be removed from the factor with only marginal effects on the Cronbach’s alpha value
statistic.
Similarly, item 25 indicated a high cross loading with Factor 3 (management
by exception active) and Factor 4 (vision). Again, reliability analysis indicated that
this item could be removed with only marginal consequences for overall alpha
values.
Two items in Factor 3 (management by exception active) also indicated
problems with both cross loadings and low item loadings. Items 27 and 4 indicated a
high cross loading with Factor 5 (undefined due to low eigenvalue and poor
interpretability), and items 4 and 22 indicated low loading values. This, along with a
poor Cronbach’s alpha statistic, suggested the factor should be dropped from the
analysis.
Items 8 and 11 in the last factor, vision, also demonstrated a high cross
loading on Factor 1 (individualised consideration) and Factor 2 (laissez-faire). Final
factor descriptions and Cronbach’s alphas are indicated in Table 5.1.8 below.
364
Table 5.1.8
Redefined Leadership Factors, Loadings and Cronbach’s alpha
Item Description Loading Alpha Factor 1—Individualised Consideration (IC) .93 31. Helps me to develop my strengths. 0.88 30. Gets me to look at problems from many different angles. 0.79 10. Instills pride in me for being associated with him/her. 0.78 21. Acts in ways that builds my respect. 0.75 1. Provides me with assistance in exchange for my efforts. 0.75 32. Suggests new ways of looking at how to complete assigned tasks.
0.75
19. Treats me as an individual rather than just a member of the group.
0.70
18. Goes beyond self-interest for the good of the group. 0.68 29. Considers me as having different needs, abilities and aspirations from others.
0.67
16. Makes clear what one can expect to receive when performance goals are achieved.
0.59
15. Spends time teaching and coaching. 0.54 Factor 2— Laissez-Faire Leadership (LF) .86 12. Waits for things to go wrong before taking action. 0.77 20. Demonstrates that problems must become chronic before taking action.
0.73
28. Avoids making decisions. 0.70 33. Delays responding to urgent questions. 0.69 7. Is absent when needed. 0.64 5. Avoids getting involved when important issues arise. 0.62 3. Fails to interfere until problems become serious. 0.60 Factor 4—Vision (VI) .86 14. Specifies the importance of having a strong sense of purpose. -0.79 13. Talks enthusiastically about what needs to be accomplished. -0.73 9. Talks optimistically about the future. -0.71 34. Emphasises the importance of having a collective sense of mission.
-0.67
26. Articulates a compelling vision for the future. -0.67 36. Expresses confidence that goals will be achieved. -0.67 6. Talks about their most important values and beliefs regarding education.
-0.46
365
Confirmatory factor analysis
Confirmatory factor analysis (CFA) is a statistical technique that determines
if the number of factors and the loadings of indicator variables on them conform to
what is expected on the basis of a researcher’s à priori assumptions. Indicator
variables are selected on the basis of prior theory, and confirmatory factor analysis is
applied to determine if observed variables (items) load as predicted on the expected
factors.
An advantage of confirmatory factor analysis is that it allows the testing of
hypotheses regarding specific factor structures. Therefore, the null hypothesis is
specified first. All confirmatory factor analysis was undertaken using LISREL 8.54
(Jöreskog & Sörbom, 2003), using a maximum likelihood method. A detailed
presentation of the conduct of a confirmatory factor analysis is presented in Chapter
4, and can be found elsewhere (e.g., Bollen, 1989; Byrne, 1998; Jöreskog & Sörbom,
1993; Marsh, Balla, & Hau, 1996; Marsh, Balla, & McDonald, 1988).
In a CFA, a researcher posits an à priori structure and then tests a solutions
ability to fit the data. Parameters usually employed include factor loadings, factor
variances and covariances, and measured variable uniquenesses. Common measures
used to evaluate goodness of fit include the Tucker-Lewis index (TLI), the relative
noncentrality index (RNI), and the root mean square error of approximation
(RMSEA), along with the Normal Theory χ2 test statistic (the default in LISREL)
and an evaluation of parameter estimates (Marsh, Balla & Hau, 1996; Marsh, Balla
& McDonald, 1988; Marsh & Hau, 2004). Both the TLI and the RNI will range
along a 0 to 1 continuum, with values greater than .90 and .95 considered to
represent acceptable to excellent fits to the data. RMSEA values between .05 and .08
indicate a close fit, where values between .08 and .10 reflect a mediocre fit. RMSEA
values greater than .10 are not considered acceptable. The RNI does not penalise a
solution for lack of parsimony, whereas the TLI and RMSEA contains penalties for a
lack of parsimony.
Measurement Model. The exploratory factor analysis for the structure of
leadership (see previous section) initially suggested the data supported a four-factor
366
leadership model. This model is based on four latent variables: individualised
consideration (IC) measured by thirteen observed variables, laissez-faire leadership
(LF) measured by eight observed variables, management by exception—active
(MBEA) measured by four observed variables, and vision (VI) measured by nine
observed variables.
The MLQ-5X (Short) developed by Bass and Avolio (1997) is based on nine
latent variables, each measured by four observed variables (see Table 5.1.1). The
latent variables include idealised influence (behaviour), idealised influence
(attributed), inspirational motivation, intellectual stimulation, individualised
consideration, contingent reward, management by exception—active, management
by exception —passive and laissez-faire.
Model Specification. It is customary in most applications of confirmatory
factor analysis to hypothesise that the latent variables are caused by the observed
variables. All the confirmatory factor analyses undertaken as part of this study are
said to be recursive because a causal flow is expected from the latent variables to the
observed variables.
Model Identification. Under normal circumstances, the standard method of
producing an asymptotic covariance matrix when the data is ordinal is to employ a
weighted least squares methodology. However, if the sample size is too small to
produce an asymptotic covariance matrix, Jöreskog and Sörbom (1996) suggest that
the covariance matrix be produced using a maximum likelihood estimation method.
This was done in all the confirmatory factor models developed as part of this study,
using LISREL 8.54 (Jöreskog and Sörbom, 2003).
Model Estimation. Fit statistics based on the analysis of the data set using
Bass et al.’s (1997) nine factor model is shown in comparison with the four factor
and three factor models below in Table 5.1.10. An initial analysis suggested a poor
fit between the model proposed by Bass et al. (1997) and the data, with several items
recording low squared multiple correlation (R2) statistics. A small R2 indicates a
weak relation and is suggestive that the item is not an effective measure (Jöreskog &
Sörbom, 1993).
367
Further, correlation statistics indicated that several of Bass et al.’s (1997)
factors were highly correlated with each other, suggesting that either the original
factor is poorly defined or that the items within these factors failed to adequately
discriminate these factors from other factors. The correlations between Bass et al.’s
(1997) original nine factors are shown below in Table 5.1.9.
An initial analysis of the four-factor model indicated high cross loading with
items 24 (“keeps track of all mistakes”), 22 (“concentrates his/her full attention on
dealing with mistakes, complaints and failures”) and 27 (“directs my attention
towards failures to meet standards”) on each of the other three factors. This, along
with low squared multiple correlation (R2) measures indicated that the factor
management by exception active was not effectively discriminated in the data set.
The result was the model was further reduced to a three-factor model.
Table 5.1.9
Nine factor correlations (Italicised—levels of significance)
Idea
lised
Influ
ence
(B
ehav
iour
)
Idea
lised
Influ
ence
(A
ttrib
ute
d)
Insp
iratio
nal
Mo
tivat
ion
Inte
llect
ual
Stim
ula
tion
Ind
ivid
ua
lised
Co
nsi
de
ratio
n
Co
ntin
gen
t Rew
ard
Man
ag
eme
nt b
y E
xcep
tion
(A
ctiv
e)
Man
ag
eme
nt b
y E
xcep
tion
(P
assi
ve)
Lais
sez-
Fai
re
Idealised Influence (Behaviour) 1.00
.
Idealised Influence (Attributed) 0.83 1.00
0.02 .
Inspirational Motivation 0.94 0.82 1.00
0.02 0.02 .
Intellectual Stimulation 0.80 0.88 0.77 1.00
0.03 0.02 0.03 .
Individualised Consideration 0.69 0.90 0.68 0.92 1.00
0.03 0.02 0.03 0.02 .
Contingent Reward 0.84 0.96 0.81 0.91 0.96 1.00
0.03 0.01 0.03 0.02 0.02 .
Management by Exception (Active) 0.01 -0.13 -0.06 -0.01 -0.19 -0.09 1.00
0.06 0.06 0.06 0.06 0.06 0.06 .
Management by Exception (Passive)-0.56 -0.65 -0.58 -0.51 -0.54 -0.60 -0.06 1.00
0.04 0.03 0.04 0.04 0.04 0.04 0.06 .
Laissez-Faire -0.61 -0.69 -0.58 -0.58 -0.59 -0.72 -0.02 -0.88 1.00
0.04 0.03 0.04 0.04 0.04 0.03 0.06 0.02 .
368
Confirmatory factor analysis on the three factors of vision (VI),
individualised consideration (IC), and laissez-faire leadership (LF) produced fit
statistics as indicated in Table 5.1.10. While this model produced an acceptable
solution, an analysis of the fit statistics suggested that model trimming and building
could further improve several of the indicated statistics.
Table 5.1.10
Fit indices for models
Model Nine-factor Model
(Bass et al., (1997)
Four-factor Model
(EFA Model)
Three-factor model
Three-factor model
(respecified)
χ2a 2635.011 2633.752 1343.446 321.229 dfb 558 521 272 101 pc 0.0 0.0 0.0 0.0 RMSEAd .0902 .0942 .0928 .0691 TLI e .946 .945 .959 .971 RNI f .953 .949 .963 .975 SMSRg .0811 .0817 .0567 .0486 Note: aχ2 – Chi Square bdf – Degrees of Freedom cp – Probability Level dRMSEA—Root Mean Square Error of ApproximationeTLI—Tucker-Lewis Index fRNI—Relative Noncentrality Index
gSMSR—Standardised Root Mean Square Residual
Assessment of Fit. Confirmatory factor analysis of the remaining three-factor
model produced an interpretable solution with good fit parameter statistics. A
comparison between Bass and Avolio’s (1997) nine-factor model and the three-factor
model is presented in Table 5.1.10. While the Root Mean Square Error of
Approximation (RMSEA) rose by 0.0026, all other fit parameter indicated the three-
factor model was a more appropriate solution given the data set. This is evident in
the results obtained from the Tucker Lewis Index (TLI; 0.959) and Relative
Noncentrality Index (RNI; 0.963), both of which indicated an improvement in fit
over the nine-factor and four-factor models.
369
It must be noted, however, that the three-factor model’s RMSEA fit
parameter index still lies outside what is considered acceptable. Marsh & Hau, 2004,
suggested that for the TLI and RNI, fit statistic results of 0.90 and 0.95 respectively
and above are generally regarded as acceptable, the closer to one the better. Further,
Browne and Cudeck (1993) argued that RMSEA statistics of 0.05 or less indicates a
close fit to the data. Finally, a Standardised Root Mean Square Residual (SMSR)
statistic of between .05 and .06 is thought to indicate good model fit (Cliff, 1983).
While the three-factor model’s fit parameter indices are more appropriate
than the nine-factor model’s, the best that can be said is that the hypothesised three-
factor model is a more appropriate solution given the data set than Bass and Avolio’s
(1997) nine-factor model.
Model Re-Specification. Given the three-factor model’s fit parameter indices,
it was necessary to apply model building and trimming techniques to explore the
possibility of improved model fit. An examination of parameter estimates suggested
that several items contributed only marginally to their respective latent variables, and
therefore could be trimmed. Fit statistics from the respecified three-factor model are
also given in Table 5.1.10. Key indices fall in appropriate ranges, and the respecified
model proposed a good solution given the data set.
The correlation between latent variables is shown in Table 5.1.10. Correlation
statistics are all less than one, and range between -.55 and .69, and none of the
factors correlating highly with the other two factors.
Table 5.1.10
Correlation Matrix of the Respecified Three-Factor Leadership Model
VI IC LF Vision (VI)
1.00
.
Individualised Consideration (IC)
0.69 0.03
1.00 .
Laissez-Faire Leadership(LF)
-0.55 0.04
-0.63 0.04
1.00 .
370
The observed variable response items and descriptors for the leadership
scales from the fitted three-factor leadership measurement model are shown below in
Table 5.1.11.
The standardised solution, parameter estimates and residuals for the final
fitted three-factor leadership measurement model are shown below in Figure 5.1.2. A
chi square value of 321.229 (df = 101) was obtained for this solution, and the
accompanying fit statistics indicated a very good fit (TLI = .971; RNI = .975;
RMSEA = 0.0691). All these fit statistics were well within the acceptable range for
being considered a good fit.
Table 5.1.11
Response items and descriptors for leadership scales
Item Descriptor Scale: Vision (VI) Cronbach’s Alpha: 0.80 Item 6—Talks about their most important values and beliefs regarding education. Item 9—Talks optimistically about the future. Item 13—Talks enthusiastically about what needs to be accomplished. Item 14—Specifies the importance of having a strong sense of purpose. Item 26—Articulates a compelling vision for the future. Scale: Individualised Consideration (IC) Cronbach’s Alpha: 0.82 Item 1—Provides me with assistance in exchange for my efforts. Item 15—Spends time teaching and coaching. Item 19—Treats me as an individual rather than just a member of the group. Item 29—Considers me as having different needs, abilities and aspirations from others. Item 35—Expresses satisfaction when I meet expectations. Scale: Laissez-Faire Leadership (LF) Cronbach’s Alpha: 0.83 Item 3—Fails to interfere until problems become serious. Item 5—Avoids getting involved when important issues arise. Item 7—Is absent when needed. Item 12—Waits for things to go wrong before taking action. Item 28—Avoids making decisions. Item 33—Delays responding to urgent questions.
371
0.82 M6 0.42
0.36 M9 0.80
0.29 M13 0.84 VI 1.00
0.35 M14 0.81
0.45 M26 0.74
0.69
0.30 M1 0.84
0.56 M15 0.60
0.31 M19 0.83 IC 1.00
0.57 M29 0.66 -0.55
0.38 M35 0.79
0.56 M3 0.66 -0.63
0.42 M5 0.76
0.55 M7 0.67
LF 1.00
0.39 M12 0.78
0.41 M28 0.77
0.45 M33 0.74
Chi-Square = 321.23; df = 101; P-value = 0.000; RMSEA = 0.069
Figure 5.1.2. Standardised solution to fitted three-factor leadership measurement
model.
School Learning Environment
Exploratory factor analysis
The School Level Environment Questionnaire (SLEQ) consisted of 56 items
designed to examine eight school climate constructs, including student
supportiveness (SS), affiliation (AFF), professional interest (PI), achievement
orientation (AO), centralisation (CEN), innovativeness (INN), resource adequacy
(RA) and formalisation (FOR). An analysis of these 56 items was undertaken using
principal axis factoring with oblimin rotation. Several items were negatively worded
and required reverse scoring. Seven identifiable factors were produced with
372
eigenvalues greater than 1, the results of which are presented in Table 5.1.12. The
scree plot is shown as Figure 5.1.3 below.
Table 5.1.12
SLEQ 7 factor Eigenvalues, Variation and Cumulative Percentage
Factor Number
Eigenvalues Percent of Variation
Cumulative Percentage
1 8.51 15.2 15.2 2 3.58 6.4 21.6 3 2.66 4.7 26.3 4 2.07 3.7 30.0 5 1.70 3.0 33.1 6 1.52 2.7 35.8 7 1.00 1.8 37.6
Scree Plot
Component Number
55524946434037343128252219161310741
Eig
enva
lue
10
8
6
4
2
0
Figure 5.1.3. School Level Environment Questionnaire Scree Plot.
While this factor analysis did not produce the clean, eight factor structure
proposed by Fisher and Fraser (1991), a seven factor structure was supported by the
data, as shown in the pattern matrix in Table 5.1.13. A structure matrix showing
cross loadings is displayed below in Table 5.1.14.
373
Table 5.1.13
SLEQ Pattern Matrix Factors
Item Original construct 1 2 3 4 5 6 7
E42 Affiliation .69 E10 Affiliation .63 E50 Affiliation .61 E18 Affiliation .50 E34 Affiliation .44 E43 Professional Interest .41 E26 Affiliation .40 E2 Affiliation .36 E17 Student Supportiveness .79 E41 Student Supportiveness .71 E9 Student Supportiveness .69 E33 Student Supportiveness .52 E56 Achievement Orientation .76 E16 Achievement Orientation .64 E32 Achievement Orientation .58 E40 Achievement Orientation .48 E24 Achievement Orientation .45 E8 Achievement Orientation .44 E45 Centralisation -.72 E37 Centralisation -.67 E21 Centralisation -.58 E13 Centralisation -.39 E55 Resource Adequacy .81 E31 Resource Adequacy .62 E23 Resource Adequacy .56 E39 Resource Adequacy .51 E7 Resource Adequacy .35 E15 Resource Adequacy E47 Resource Adequacy E53 Centralisation .70 E5 Centralisation .70 E29 Centralisation .69 E6 Innovativeness .60 E14 Innovativeness .36 E54 Innovativeness .70 E46 Innovativeness .70 E27 Professional Interest E11 Professional Interest E3 Professional Interest E1 Student Supportiveness E25 Student Supportiveness E49 Student Supportiveness E44 Formalisation E20 Formalisation E52 Formalisation E35 Professional Interest E30 Innovativeness E22 Innovativeness E48 Achievement Orientation E4 Formalisation E12 Formalisation E36 Formalisation E38 Innovativeness E19 Professional Interest E51 Professional Interest E28 Formalisation Extraction Method: Principal Axis Analysis Rotation Method: Oblimin
374
Convergence: 39 iterations Table 5.1.14
SLEQ Structure Matrix Factors
Original construct 1 2 3 4 5 6 7
E42 Affiliation .72 E50 Affiliation .67 E10 Affiliation .65 E18 Affiliation .62 E34 Affiliation .60 -.31 .38 .30 E43 Professional Interest .54 .33 E2 Affiliation .51 .37 .31 E26 Affiliation .51 E17 Student Supportiveness .83 E41 Student Supportiveness .81 E9 Student Supportiveness .73 E33 Student Supportiveness .61 E56 Achievement Orientation .72 E16 Achievement Orientation .65 E32 Achievement Orientation .59 E40 Achievement Orientation .51 E24 Achievement Orientation .49 E8 Achievement Orientation .47 E45 Centralisation -.74 E37 Centralisation -.70 E21 Centralisation -.60 E13 Centralisation -.51 .33 E55 Resource Adequacy .80 E23 Resource Adequacy .62 E31 Resource Adequacy .60 E39 Resource Adequacy .58 E15 Resource Adequacy .42 E7 Resource Adequacy .38 E47 Resource Adequacy .34 .33 E53 Centralisation .39 .78 .35 E6 Centralisation .72 .38 E5 Centralisation .69 E29 Innovativeness .69 .36 E14 Innovativeness -.31 .56 .50 E19 Innovativeness .41 E54 Innovativeness .36 .74 E46 Professional Interest .70 E51 Professional Interest .34 .32 .49 E27 Professional Interest .31 .40 E11 Student Supportiveness .41 .31 E3 Student Supportiveness .38 E35 Student Supportiveness E1 Formalisation .52 E25 Formalisation .47 E49 Formalisation .40 E44 Professional Interest E52 Innovativeness E20 Innovativeness E30 Achievement Orientation E36 Formalisation -.43 E12 Formalisation E4 Formalisation E38 Innovativeness .37 E22 Professional Interest .37 .33 E48 Professional Interest E28 Formalisation Extraction Method: Principal Axis Analysis Rotation Method: Oblimin Convergence: 39 iterations
375
A correlation matrix is presented in Table 5.1.15 that indicates correlation co-
efficients range from between -.26 and .37 between the seven identified factors.
Table 5.1.15
SLEQ Factor Correlation Matrix
Factors 1 2 3 4 5 6 7
1 1.00 2 .25 1.00 3 .03 .07 1.00 4 -.26 -.20 -.00 1.00 5 .13 .14 -.15 -.05 1.00 6 .27 .15 -.11 -.20 .17 1.00 7 .23 .07 .00 -.11 .21 .37 1.00
Extraction Method: Principal Axis Factoring Rotation Method: Oblimin
Reliability analysis based on Fisher and Fraser’s (1991) eight-factor model
produced Cronbach’s alpha co-efficients as indicated in Table 5.1.16 below. This
was again suggestive that there was evidence of an underlying factor structure that
confirmatory factor analysis should clarify.
Table 5.1.16
School Level Environment Questionnaire Factors and Cronbach’s Alphas
Item Description Loadings Alpha Factor 1—Student Supportiveness .87 1. There are many disruptive, difficult students in the School. .78 9. Most students are helpful and co-operative to teachers. .65 17. Most students are pleasant and friendly to teachers. .76 25. There are many noisy, badly behaved students. .71 33. Students get along well with teachers. .61 41. Most students are well-mannered and respectful to the School staff.
.83
49. Very strict discipline is needed to control many of the students.
.61
Factor 2—Affiliation .81 2. I seldom receive encouragement from colleagues. .57 10. I feel accepted by other teachers. .60 18. I am ignored by other teachers. .65
376
26. I feel I could rely on my colleagues for assistance if I should need it.
.51
34. My colleagues seldom take notice of my professional views and opinions.
.66
42. I feel that I have many friends among my colleagues at this School.
.66
50. I often feel lonely and left out of things in the staffroom. .70 Factor 3—Professional Interest .72 3. Teachers frequently discuss teaching methods and strategies with each other.
.61
11. Teachers avoid talking with each other about teaching and learning.
.59
19. Professional matters are rarely discussed at staff meetings. .38 27. Many teachers attend inservice and other professional development courses.
.33
35. Teachers show little interest in what is happening in other schools.
.54
43. Teachers are keen to learn from their colleagues. .64 51. Teachers show considerable interest in the professional activities of their colleagues.
.67
Factor 4—Achievement Orientation .73 8. There is constant pressure to keep working. .48 16. Teachers have to work long hours to complete their work. .63 24. Teachers don’t have to work very hard in this School. .50 32. There is not time for teachers to relax. .54 40. You can take it easy and still get the work done. .54 48. Seldom are there deadlines to be met. .30 56. It is hard to keep up with your workload. .67 Factor 5—Formalisation .44 4. I am often supervised to ensure that I follow directions correctly.
.21
12. I am not expected to conform to a particular teaching style. .26 20. It is considered very important that I closely follow syllabuses and lesson plans.
.42
28. There are a few rules and regulations that I am expected to follow.
.14
36. I am allowed to do almost as I please in the classroom. .42 44. My classes are expected to use prescribed textbooks and prescribed resource material.
.42
52. I am expected to maintain very strict control in the classroom.
.38
377
Factor 6—Centralisation .75 5. Decisions about running the School are often made by the Principal or a small group of teachers.
.51
13. I have to refer even small matters to a senior member of staff for a final answer.
.57
21. Actions can usually be taken without gaining the approval of the subject department head or a senior member of staff.
.49
29. Teachers are frequently asked to participate in decisions concerning administrative policies and procedures.
.46
37. I am encouraged to make decisions without reference to a senior member of staff.
.61
45. I must ask my subject department head or senior member of staff before I do most things.
.59
53. I have very little say in the running of the School. .65 Factor 7—Innovation .75 6. It is very difficult to change anything in this School. .67 14. Teachers are encouraged to be innovative in this School. .64 22. There is a great deal of resistance to proposals for curriculum change.
.56
30. Most teachers like the idea of change. .33 38. New courses or curriculum materials are seldom implemented in the School.
.45
46. There is much experimentation with different teaching approaches.
.53
54. New and different ideas are always being tried out in this School.
.67
Factor 8—Resource Adequacy .73 7. The School or Department library includes an adequate selection of books and periodicals.
.39
15. The supply of equipment and resources is inadequate. .46 23. Video equipment, tapes and films are ready available and accessible.
.64
31. Adequate duplicating facilities and services are available to teachers.
.55
39. Tape recorders and cassettes are seldom available when needed.
.58
47. Facilities are inadequate for catering for a variety of classroom activities and learning groups of different sizes.
.40
55. Video’s, overhead transparencies and access to photocopying are usually available when needed.
.77
378
Confirmatory Factor Analysis
Measurement Model. An initial exploratory factor analysis of the School
Level Environment Questionnaire (SLEQ; see above) supported an underlying eight-
factor model, although the items that were extracted on some of the constructs were
misplaced. The eight latent variables, described by Fisher and Fraser (1991)
included: student supportiveness (SS), affiliation (AFF), professional interest (PI),
achievement orientation (AO), formalisation (FOR), centralisation (CEN),
innovation (INN) and resource adequacy (RA). Each latent variable is measured by
seven observed variables.
Assessment of Fit. A comparison between the two proposed models is
shown below in Table 5.1.17.
Table 5.1.17
Fit indices for eight factor and seven factor respecified models
Model Eight-factor Model
Seven-factor Model (EFA Model)
Seven-factor Model (respecified)
χ2a 5944.434 1600.001 638.276 dfb 1456 474 254 pc 0.0 0.0 0.0 RMSEAd .0821 .0721 .0575 TLI e .843 .908 .939 RNIf .852 .918 .949 SMSRg .0809 .0705 .0575 Note: aχ2 – Chi Square bdf – Degrees of Freedom cp – Probability Level dRMSEA—Root Mean Square Error of Approximation eTLI—Tucker-Lewis Index fRNI—Relative Non-centrality Index
gSMSR—Standardised Root Mean Square Residual
The confirmatory factor analysis of the model produced an interpretable
solution supported by good correlational discrimination between the eight factors
(see Table 5.1.18 below). However, an analysis of the fit statistics suggested a poor
model fit, with RMSEA (0.0821) well above the accepted 0.05 level, and TLI (0.843)
379
and RNI (0.852) well below the accepted 0.90 level (Marsh & Hau, 2004). SMSR
(0.0809) was also reported outside the acceptable 0.05 to .0.06 range (Cliff, 1983). A
seven factor, EFA suggested model produced better fit indices than Fisher and
Fraser’s (1991) model, with TLI and RNI indices increasing to .908 and .918
respectively. The RMSEA index also improved over the values obtained in the first
model (.0721), but was still outside the acceptable range.
Given the theoretical and conceptual underpinnings of the Fisher and Fraser
(1991) model, the decision was taken to retain the original eight-factor model.
However, model trimming and building techniques were used to improve fit
statistics.
Table 5.1.18
Eight factor school learning environment correlations (Italicised—levels of significance)
Stu
den
t Sup
por
tiven
ess
(S
S)
Affi
liatio
n (
AF
F)
Pro
fess
iona
l Int
eres
t (P
I)
For
mal
isat
ion
(F
OR
)
Cen
tral
isat
ion
(CE
N)
Inno
vatio
n (I
NN
)
Res
ourc
e A
deq
uacy
(R
A)
Ach
ieve
men
t Orie
nta
tion
(
AO
) Student Supportiveness (SS) 1.00
.
Affiliation (AFF) 0.41 1.00
0.04 .
Professional Interest (PI) 0.24 0.77 1.00
0.05 0.03 .
Formalisation (FOR) 0.45 0.44 0.05 1.00
0.06 0.06 0.07 .
Centralisation (CEN) 0.32 0.67 0.59 0.53 1.00
0.05 0.04 0.04 0.06 .
Innovation (INN) 0.24 0.56 0.78 0.13 0.82 1.00
0.05 0.04 0.03 0.07 0.03 .
Resource Adequacy (RA) 0.39 0.34 0.26 0.22 0.42 0.39 1.00
0.05 0.05 0.05 0.07 0.05 0.05 .
Achievement Orientation (AO) 0.09 0.10 0.08 0.00 0.00 0.03 -0.07 1.00
0.05 0.06 0.06 0.07 0.06 0.06 0.06 .
380
Model re-specification. An examination of the fit statistics for the eight-
factor model indicated that model trimming and building techniques (Kelloway,
1998) could be used to improve model fit. Several items recorded low squared
multiple correlation (R2) values, suggesting a weak relation between the item and the
latent variable it measured (Jöreskog & Sörbom, 1993). Further, several items
indicated a strong tendency to cross load on multiple factors, again suggestive of
those items’ poor ability to discriminate between factors.
The factor formalisation (FOR) had several items with very low squared
multiple correlation (R2) values, and the remaining items cross loaded on other
factors, particularly centralisation (CEN). Model modification indices indicated
removal of the factor completely as a step to improve fit, the result being a seven-
factor respecified model. Table 5.1.17 shows the fit statistics for the respecified
seven-factor model, while Table 5.1.19 shows the correlation indices of the
Table 5.1.19
Seven factor school learning environment model correlations (Italicised—levels of significance)
Stu
den
t
Sup
port
iven
ess
(SS
)
Affi
liatio
n (
AF
F)
Pro
fess
iona
l Int
eres
t
(P
I)
Cen
tral
isat
ion
(C
EN
)
Inno
vatio
n (
INN
)
Res
ourc
e A
dequ
acy
(RA
)
Ach
ieve
men
t
Orie
nta
tion
(A
O)
Student Supportiveness (SS) 1.00
.
Affiliation (AFF) 0.44 1.00
0.05 .
Professional Interest (PI) 0.20 0.60 1.00
0.06 0.05 .
Centralisation (CEN) 0.27 0.48 0.49 1.00
0.05 0.05 0.05 .
Innovation (INN) 0.22 0.47 0.77 0.81 1.00
0.06 0.05 0.04 0.04 .
Resource Adequacy (RA) 0.42 0.44 0.27 0.41 0.41 1.00
0.05 0.05 0.06 0.05 0.06 .
Achievement Orientation (AO) 0.02 0.01 -0.01 -0.08 -0.03 -0.07 1.00
0.06 0.06 0.06 0.06 0.06 0.06 .
381
remaining seven-factor model. Table 5.1.20 outlines item descriptions and associated
Cronbach’s alpha values. RMSEA fit statistics improved greatly with the respecified
seven-factor model, dropping from 0.0821 to 0.0575 and within the range of
values generally seen as acceptable (Browne & Cudeck, 1993). TLI and RNI fit
statistics also showed improvement, increasing from 0.843 to 0.939 and from 0.852
to 0.949 respectively. Lastly, the SMSR statistic also indicated improvement in fit
between the eight and seven-factor (respecified) model, decreasing from 0.0809 to
0.0575, and again, within the acceptable range (Cliff, 1983). Correlation indices
ranged between 0.81 and –0.08, with the majority of correlations occurring at the 0.4
level. Cronbach’s alpha, which is a commonly used measure of reliability (Nunnally,
1978), ranged from between 0.79 and 0.61, with four of the seven scales used in the
analysis above the 0.70 level, and another two scales indicating reliability values of
0.69 and 0.67. One factor, innovation, recorded a Cronbach’s alpha of 0.61.
Table 5.1.20
School learning environment factors and response item descriptions
Item Descriptor Scale: Student Supportiveness (SS) Cronbach’s Alpha: 0.79 Item 17—Most students are pleasant and friendly to teachers. Item 25—There are many noisy, badly behaved students. Item 33—Students get along well with teachers. Item 41—Most students are well-mannered and respectful to the School staff. Scale: Affiliation (AFF) Cronbach’s Alpha: 0.73 Item 10—I feel accepted by other teachers. Item 26—I feel I could rely on my colleagues for assistance should I need it. Item 42—I feel that I have many friends among my colleagues at this school. Item 50—I often feel lonely and left out of things in the staffroom. Scale: Professional Interest (PI) Cronbach’s Alpha: 0.67 Item 3—Teachers frequently discuss teaching methods and strategies with each other. Item 11—Teachers avoid talking to each other about teaching and learning. Item 51—Teachers show considerable interest in the professional activities of their colleagues. Scale: Achievement Orientation (AO) Cronbach’s Alpha: 0.69 Item 16—Teachers have to work long hours to complete their work. Item 32—There is not time for teachers to relax. Item 40—You can take it easy and still get the work done. Item 56—It is hard to keep up with your workload.
382
Scale: Centralisation (CEN) Cronbach’s Alpha: 0.71 Item 5—Decisions about running the School are often made by the Principal or a small group of teachers. Item 29—Teachers are frequently asked to participate in decisions concerning administrative policies and procedures. Item 37—I am encouraged to make decisions without reference to a senior member of Staff Item 53—I have very little say in the running of the School. Scale: Innovation (INN) Cronbach’s Alpha: 0.61 Item 14—Teachers are encouraged to be innovative in this school Item 22—There is a great deal of resistance to proposals for curriculum change. Item 54—New and different ideas are always being tried out in this school. Scale: Resource Adequacy (RA) Cronbach’s Alpha: 0.71 Item 23—Video equipment, tapes and films are readily available and accessible. Item 39—Tape recorders and cassettes are seldom available when needed. Item 55—Videos, overhead transparencies and access to photocopying are usually available when needed
383
0.22 E17 0.88
0.56 E25 0.66
SS 1.00
0.48 E33 0.72
0.14 E41 0.93
0.46 E10 0.74
0.56 E26 0.55
AFF 1.00
0.42 E42 0.75
0.47 E50 0.73
0.53 E3 0.68
0.42 E11 0.76 PI 1.00
0.63 E51 0.50
0.53 E5 0.69
0.54 E29 0.68 CEN 1.00
0.83 E37 0.41
0.27 E53 0.85
0.40 E14 0.78
0.70 E22 0.55 INN 1.00
0.67 E54 0.58
0.48 E23 0.72
0.49 E39 0.71 RA 1.00
0.41 E55 0.77
0.52 E16 0.69
0.59 E32 0.64
AO 1.00
0.77 E40 0.48
0.34 E56 0.81
Chi-Square = 638.28; df = 254; P-value = 0.000; RMSEA = 0.058
Figure 5.1.4. Standardised solution for school learning environment model.
384
The standardised solution and residuals for the fitted seven factor school
learning environment measurement model are shown in Figure 5.1.4, while
parameter estimates are given in Table 5.1.21.
Table 5.1.21
Standardised parameter estimates, fitted seven factor school learning environment model SS AFF PI CEN INN RA AO Student Supportiveness (SS) 0.44 0.21 0.27 0.22 0.42 0.02 Affiliation (AFF) 0.44 0.60 0.48 0.47 0.43 0.01 Professional Interest (PI) 0.21 0.60 0.49 0.77 0.27 -0.01 Centralisation (CEN) 0.27 0.48 0.49 0.81 0.40 -0.08 Innovation (INN) 0.22 0.47 0.77 0.81 0.41 -0.03 Resource Adequacy (RA) 0.42 0.43 0.27 0.40 0.41 -0.07 Achievement Orientation (AO)
0.02 0.01 -0.01 -0.08 -0.03 -0.07
Teacher Outcomes
Introduction
Four separate, composite outcomes measures were formulated using items
from the MLQ-5X (Short) (Bass & Avolio, 1997), School Level Environment
Questionnaire (Fisher & Fraser, 1991) and Patterns of Adaptive Learning Survey
(Midgley, Maehr, Hicks, Roeser, Urdan, Anderman & Kaplan, 1996). These outcome
measures were designed to examine teacher’s perceptions of global satisfaction with
leadership, influence, control and effectiveness. These outcomes were measured
using the Teacher’s Satisfaction Questionnaire (Appendix 4.3).
Exploratory factor analysis
Teacher Perceptions of Global Satisfaction with Leadership (SAT). Items
37 to 45 on the Multifactor Leadership Questionnaire (5X- Short) reportedly
measured the three factors of motivation to extra effort, perceptions of effectiveness
of the leader and global satisfaction with the leader. An exploratory factor analysis,
using principal-axis extraction, was undertaken on these items to examine these
constructs with the data set collected from New South Wales’ secondary schools.
385
Only one factor could be extracted from this data set. This one factor
generated an eigenvalue of 5.513, and accounted for 61.26% of the explained
variance in the data set.
This factor, global satisfaction with leadership (SAT) examines issues of
overall follower willingness to give extra effort, follower satisfaction with leadership
and follower perceptions of leadership effectiveness. The global satisfaction with
leadership items and their loadings are shown below (see Table 5.1.22).
Table 5.1.22
Global Satisfaction with Leadership, Loadings and Cronbach’s Alpha
Item Description Loading Alpha Factor 1—Teacher Satisfaction with Leadership (SAT) .93 44. Increases my willingness to try harder .87 38. Uses methods of leadership that are satisfying. .87 42. Heightens my desire to succeed. .86 37. Is effective in meeting my job related needs. .85 41. Works with me in a satisfactory way. .84 45. Leads a group that is effective. .79 40. Is effective in representing me to higher authority. .79 43. Is effective in meeting organisational requirements. .66 39. Gets me to do more than I expected to do. .38
Item 39 “gets me to do more than I expected to do” is difficult to include, as
its loading value is less than .4. Further evidence of the advantage of removing this
item comes from the reliability analysis, which suggested that if item 39 is dropped,
then Cronbach’s alpha increased to .94.
Perceptions of Teacher Influence (INFLU). This outcome measure
comprised four items that examine teachers’ overall perception of their influence in
their current school, and was made up of items taken from the Patterns of Adaptive
Learning Survey (Midgley et al. 1996). Item description and Cronbach’s alpha
statistics are shown below in Table 5.1.23.
386
Table 5.1.23
Perceptions of Teacher Influence and Cronbach’s alpha
Item Description Alpha Factor 1—Perceptions of teacher influence (INFLU) .73 4. In this school, some teachers have more influence than other teachers
.68
11. In this school, the administration shows favouritism to some teachers
.86
38. Power and influence count a lot around this school. .86
Perceptions of Teacher Effectiveness (TEFF). This factor, also
compromised of items taken from Patterns of Adaptive Learning Survey (Midgley et
al. 1996), examines teacher perceptions of their effectiveness as teachers within their
respective schools. Item descriptions and Cronbach’s alpha is shown below in Table
5.1.24.
Table 5.1.24
Perceptions of Teacher Effectiveness and Cronbach’s alpha.
Item Description Alpha Factor 1—Perceptions of teacher effectiveness (TEFF) .74 21. I can deal with almost any learning problem .73 28. I am certain I am making a difference in the lives of my students
.73
34. If I try really hard, I can get through to even the most difficult students
.79
39. I am good at helping all the students in my class make significant improvement
.75
Perceptions of Teacher Control (TCON). This factor measures the amount
of control or autonomy teachers perceive they exercise in the course of their daily
teaching. It is derived from several items that were discarded in subsequent
confirmatory factor analysis, in the process of model trimming and building. These
items were taken from the centralisation (items 13, 37 and 45) and formalisation
(items 36) factors of the School Level Environment Questionnaire (Fisher and Fraser
1991). Item descriptions and Cronbach’s alpha are indicated below in Table 5.1.25.
387
Table 5.1.25
Perception of Teacher Control and Cronbach’s alpha.
Item Description Alpha Factor 1—Teacher perception of control (TCON) .69 13. I have to refer even small matters to a senior member of staff for a final answer
.71
36. I am allowed to do almost as I please in the classroom. .58 37. I am encouraged to make decisions without reference to a senior member of staff
.75
45. I must ask my subject department head or senior member of staff before I do most things
.82
Confirmatory factor analysis
Measurement Model. Four constructs were developed to measure different
aspects of teacher job satisfaction (see above). These constructs included global
satisfaction with leadership (SAT), teacher perceptions of influence (INFLU),
teacher perceptions of effectiveness (TEFF) and teacher perceptions of control
(TCON). A confirmatory factor analysis technique, using LISREL 8.54 (Jöreskog &
Sörbom, 2003), was applied to these constructs to test the latent structure of these
constructs.
Model Specification. Items 37 to 45 of the MLQ-5X (Short) were designed to
measure three sub-constructs of follower satisfaction. These include satisfaction,
effectiveness and extra effort. The latent variable satisfaction is measured by two
observed variables, effectiveness by four observed variables and extra effort by three
observed variables.
The Patterns of Adaptive Learning Survey (Midgley, Maehr, Hicks, Roeser,
Urdan, Anderman & Kaplan, 1996, p. 2) measured several task and performance
related factors, including perceptions related to teacher efficacy, and instructional
and motivational strategies. Three items were used to measure perceptions of teacher
influence as far as the degree of influence teachers exercised in administration of
their school, and another three items to measure perceptions of teacher effectiveness.
388
A three-item scale was also developed using discarded items from the
formalisation and centralisation factors of the School Level Environment
Questionnaire (Fisher & Fraser, 1991). This scale explores teacher’s perceptions of
the locus of control or autonomy they exercise in their professional lives, and was
considered relevant given findings emanating from the qualitative phase of the study.
Assessment of fit. Table 5.1.26 below reports on the fit statistics of the
global satisfaction with leadership (SAT) component of this outcomes model.
Table 5.1.26
Fit indices for the three sub factor and single factor teacher satisfaction model
Model Three sub factor model
Single sub factor model
χ2a 152.959 314.807 dfb 24 27 pc 0.0 0.0 RMSEAd 0.108 0.153 TLI e 0.976 0.952 RNIf 0.984 0.964 SMSRg 0.0283 0.0352
Note: aχ2 – Chi Square bdf – Degrees of Freedom cp – Probability Level dRMSEA—Root Mean Square Error of Approximation e TLI—Tucker-Lewis Index f RNI—Relative Non-centrality Index
gSMSR—Standardised Root Mean Square Residual
While the TLI, RNI and SMSR fit statistics of 0.976, 0.984 and 0.0283
respectively for Bass et al.’s (1997) model seem reasonable in regard to acceptable
levels (Marsh & Hau, 2004; Cliff, 1983), the RMSEA fit statistic of 0.108 was well
outside acceptable levels (Browne & Cudeck, 1993). Further, correlations (see Table
5.1.27) among the three factors ranged from 1.00 to 0.93, indicating a perfect
correlation between two of the factors (SAT and EFF), and an extremely high degree
of correlation between the other latent factors of the model. This indicated the
solution was imperfect, and prompted the question as to whether there were three-
factors or one-factor present.
389
Table 5.1.27
Three sub factor satisfaction correlations
Satisfaction Effectiveness Extra Effort Satisfaction
SAT 1.00 .
Effectiveness
EFF 1.00 0.01
1.00 .
Extra Effort
EE 0.91 0.01
0.93 0.01
1.00 .
This sub component model was therefore respecified to reflect that only one
factor could be supported by the data. Table 5.1.26 also reports the fit statistics for
the single factor global satisfaction with leadership (SAT) component of the teacher
outcomes model. This sub component of the respecified single factor model showed
marginal change in the goodness of fit statistics over the three-factor model, with
RMSEA, TLI, RNI and SMSR values respectively indicated as 0.153, 0.952, 0.964
and 0.0352. More importantly, the problems associated with extremely high
correlations among the three latent variables disappeared by reducing this model to
one factor.
Assessment of fit—Respecified model. Table 5.1.28 shows the fit statistics
for the four-factor teacher outcome model. Three of the goodness of fit indices were
indicated at generally acceptable levels (TLI, 0.928; RNI, 0.939 and SMSR, 0.0574).
However, the RMSEA index suggested a poor fit at 0.0952, which was considered
well outside the normally accepted levels of suitability (Browne & Cudeck, 1993).
390
Table 5.1.28
Fit indices for teacher outcomes model
Model Four factor model
Four factor model (respecified)
χ2a 751.192 337.286 dfb 146 98 pc 0.0 0.0 RMSEAd 0.0952 0.0731 TLI e 0.928 0.947 RNIf 0.939 0.957 SMSRg 0.0574 0.0519 Note: aχ2 – Chi Square bdf – Degrees of Freedom cp – Probability Level dRMSEA—Root Mean Square Error of Approximation e TLI—Tucker-Lewis Index f RNI—Relative Non-centrality Index
gSMSR—Standardised Root Mean Square Residual
Model Re-Specification. While several of the fit statistics indicated a good
match between the data and the proposed model, an analysis of modification indices
was suggestive that improvements could be made to the fit of the model. This was
done by deleting item 44 (“increases my willingness to try harder”) and item 45
(“leads a group that is effective”) of the MLQ-5X (Short) questionnaire, improving
the overall fit of the model.
The fit statistics for the respecified four-factor teacher outcomes model is also
shown above in Table 5.1.28, and correlational statistics below in Table 5.1.29.
While model trimming marginally effected TLI, RNI and SMSR statistics (0.947,
0.957 and 0.0519), the RMSEA index decreased to a more acceptable level of
0.0731, indicating a much better fit of the model to the data.
The range of correlation statistics (see Table 5.1.29) was generally indicated
as being good. However, it is noted that there was a high correlation between
perceptions of effectiveness and perceptions of control (.85). Conceptual differences
between the two constructs as reflected in item descriptions (see Table 5.1.30)
suggested that these were two separate factors, and experimentation by collapsing the
391
two factors into one produced very poor indices. The decision was therefore taken to
retain the two factors, teacher perceptions of effectiveness (TEFF) and teacher
perceptions of control (TCON), as two separate factors.
Table 5.1.29
Four factor teacher outcome correlations
SAT INFLU TEFF TCON Global satisfaction with leadership (SAT)
1.00 .
Perceptions of influence (INFLU)
.23
.06 1.00
.
Perceptions of effectiveness (TEFF)
.47
.05 .13 .07
1.00
Perceptions of control (TCON)
.31
.06 .19 .08
.85
.05 1.00
.
Response items, item descriptions and Cronbach’s alphas are shown in Table
5.1.30. Cronbach’s alpha for the first three response items is indicated as being
acceptable (Global teacher satisfaction with leadership (SAT), 0.90; teacher
perceptions of influence (INFLU), 0.73; and teacher perception of effectiveness
(TEFF), 0.64). It is noted that the reliability index for the factor teacher perception of
control is low at 0.59.
Table 5.1.30
Response items and descriptors for teacher outcomes model
Item Descriptor Scale: Global teacher satisfaction with leadership (SAT) Cronbach’s Alpha: 0 .90 Item 37. Is effective in meeting my job related needs Item 38. Uses methods of leadership that are satisfying. Item 39. Gets me to do more than I expected to do. Item 40. Is effective in representing me to higher authority. Item 41. Works with me in a satisfactory way. Item 42. Heightens by desire to succeed. Item 43. Is effective in meeting organisational requirements. Scale: Perceptions of teacher influence (INFLU) Cronbach’s Alpha: 0.73 Item 4. In this school, some teachers have more influence than other teachers. Item 11. In this school, the administration shows favouritism to some teachers. Item 38. Power and influence count a lot around this school.
392
Scale: Perceptions of teacher effectiveness (TEFF) Cronbach’s Alpha: 0.64 Item 21. I can deal with almost any learning problem. Item 28. I am certain I am making a difference in the lives of my students. Item 39. I am good at helping all the students in my class make significant improvement. Scale: Perceptions of teacher control (TCON) Cronbach’s Alpha: 0.59 Item 13. I have to refer even small matters to a senior member of staff for a final answer. Item 36. I am allowed to do almost as I please in the classroom. Item 45. I must ask my subject department head or senior member of staff before I do most things.
The standardised solution and residuals for the fitted four-factor teacher
outcomes measurement model are shown in Figure 5.1.5.
0.20 M37 0.89
0.20 M38 0.89
0.84 M39 0.40
0.30 M40 0.84 SAT 1.00
0.16 M41 0.91 0.23
0.47
0.23 M42 0.88
0.31
0.51 M43 0.70
0.82 C4 0.43
0.65 C11 0.59 INFLU 1.00
0.64 C38 0.60 0.13
0.19
1.00 C21 -0.03
0.39 C28 0.78 TEFF 1.00
0.50 C39 0.71
0.85
0.40 E13 0.77
0.86 E36 0.37 TCON 1.00
0.79 E45 0.46
Chi-Square = 337.29; df = 98; P-value = 0.000; RMSEA = 0.073
Figure 5.1.5. Standardised solution for teacher outcomes model.
393
394
Correlation Analysis
Table 5.1.31 indicates the correlational co-efficient statistics between each of
the fourteen constructs considered in this study at the teacher level, along with their
level of significance. These correlations are suggestive of the differential relations
the three leadership style behaviours share with each of the school learning
environment constructs, and the teacher outcome measures. Laissez-faire leadership
(LF) has a negative correlation with each of the other constructs except achievement
orientation (AO). Individualised consideration (IC) in comparison to vision (VI) style
leadership behaviour, has a stronger relation with student supportiveness (SS),
affiliation (AFF), centralisation (CEN), innovation (INN), resource adequacy (RA),
global satisfaction with leadership (SAT), perceptions of teacher influence (INLFU)
and perceptions of teacher control (TCON). The latter result seems contrary to the
leadership literature, which suggests that visionary (VI) style behaviour should be the
leadership style behaviour that is most positively related to school learning
environment and teacher outcome constructs. A multilevel modelling technique was
employed to further explore these relations.
Summary
The analysis of the data gathered across 458 staff in 52 schools found support
for a three factor school leadership model (visionary leadership (VI), individualised
consideration (IC) and laissez-faire leadership), a seven factor school learning
environment model (student supportiveness (SS), affiliation (AFF), professional
interest (PI), centralisation (CEN), innovation (INN), resource adequacy (RA) and
achievement orientation), and a four factor teacher outcomes model (Global
satisfaction with leadership (SAT), perception of influence (INFLU), perception of
effectiveness (EFF) and perception of control (TCON). These scales that were
developed through the EFA and CFA analysis and presented in this Appendix were
used in the subsequent, multilevel analysis described in Chapter 5.
395v
r eade
rsh
each
er
.
.
.
.
.
.
Table 5.20
Factor correlations*
Vis
ion
Ind
ivid
ua
l C
onsi
der
atio
n
Lais
sez-
Fai
re
Stu
den
t S
uppo
rtiv
enes
s
Affi
liatio
n
Pro
fess
iona
l In
tere
st
Cen
tral
isat
ion
Inno
vatio
n.
Res
ourc
e A
dequ
acy
Ach
ieem
ent
Oie
nta
tion
Glo
bal
S
atis
fact
ion
with
L
ip
Per
cep
tions
of
T Influ
ence
Per
cep
tions
of
Tea
cher
E
ffect
iven
ess
Per
cep
tions
of
Tea
cher
Con
trol
Vision (VI)
1.00
Individualised Consideration (IC) 0.69 1.00 0.03 Laissez-Faire (LF) -0.56 -0.63 1.00 0.04 0.04Student Supportiveness (SS) 0.10 0.13 -0.09 1.00 0.05 0.05 0.05Affiliation (AFF) 0.23 0.35 -0.20 0.44 1.00 0.05 0.05 0.05 0.05Professional Interest (PI) 0.36 0.34 -0.35 0.21 0.61 1.00 0.05 0.05 0.05 0.06 0.05Centralisation (CEN) 0.38 0.64 -0.35 0.27 0.49 0.48 1.00
0.05 0.04 0.05 0.05 0.05 0.05 .Innovation (INN) 0.46 0.52 -0.48 0.23 0.47 0.77 0.80 1.00 0.05 0.05 0.05 0.06 0.05 0.04 0.04 .Resource Adequacy (RA) 0.07 0.20 -0.12 0.42 0.44 0.27 0.41 0.40 1.00
0.06 0.05 0.06 0.05 0.05 0.06 0.05 0.06 .Achievement Orientation (AO) -0.11 -0.11 0.20 0.02 0.01 -0.01 -0.07 -0.03 -0.07 1.00
0.06 0.06 0.05 0.05 0.06 0.06 0.06 0.06 0.06 .Global Satisfaction with Leadership 0.70 0.95 -0.70 0.10 0.30 0.37 0.60 0.56 0.19 -0.19 1.00
(SAT) 0.03 0.01 0.03 0.05 0.05 0.05 0.04 0.04 0.05 0.05 .Perception of Teacher Influence 0.37 0.43 -0.44 0.17 0.21 0.42 0.52 0.64 0.25 -0.14 0.46 1.00(INFLU) 0.05 0.05 0.05 0.06 0.06 0.06 0.05 0.05 0.06 0.06 0.06 .Perception of Teacher Effectiveness 0.37 0.34 -0.44 -0.03 0.11 0.40 0.43 0.62 0.22 -0.08 0.43 0.96 1.00
(EFF) 0.06 0.06 0.06 0.06 0.07 0.07 0.06 0.07 0.07 0.07 0.06 0.07 . Perceptions of Teacher Control 0.20 0.29 -0.10 0.28 0.50 0.34 0.52 0.54 0.34 0.00 0.21 0.13 0.03 1.00(CON) 0.06 0.05 0.06 0.05 0.05 0.06 0.05 0.05 0.06 0.06 0.05 0.06 0.07 .
* Note these correlation results are based on analysis at the teacher level rather than at the school level.
(Italicised—levels of significance)
39
5
APPENDIX 5.2
DEMOGRAPHIC CHARACTERISTICS OF
PARTICIPANTS, STUDY 1
Table 5.2.1
Response Rates Based Upon Demographic Characteristics
Basic Demographics Number Teachers 458 Schools 52 Regions 32
396
Table 5.2.2
School Distribution by District
District Number of Schools Approximate Percentage Albury 2 3.8 Armidale 1 1.9 Bathurst 2 3.8 Bateman’s Bay 2 3.8 Blacktown 3 5.8 Bankstown 1 1.9 Bondi 2 3.8 Campbelltown 1 1.9 Clarence/Coffs Harbour 2 3.8 Deniliquin 2 3.8 Fairfield 2 3.8 Granville 1 1.9 Griffith 1 1.9 Hornsby 3 5.8 Lake Macquarie 1 1.9 Lismore 1 1.9 Liverpool 2 3.8 Maitland 3 5.8 Moree 1 1.9 Mt. Druitt 2 3.8 Newcastle 1 1.9 Northern Beaches 3 5.8 Orange 1 1.9 Penrith 1 1.9 Port Macquarie 1 1.9 Queanbeyan 1 1.9 Shellharbour 1 1.9 St George 2 3.8 Sutherland 2 3.8 Tamworth 1 1.9 Wagga Wagga 1 1.9 Wollongong 2 3.8 Total 52 100.0
397
Table 5.2.3
Teacher’s School Size (Number of Students)
School Size—Student Population
Number Approximate Percentage
Less than 200 0 0.0 201 – 400 40 8.7 401 – 600 38 8.3 601 – 800 99 21.6 801 – 1000 128 27.9 1001 – 1200 101 22.1 More than 1201 27 5.9 Missing 25 5.5 Total 458 100.0
Table 5.2.4
Breakdown of Participating Teachers’ Current Position in School
Position Number Approximate Percentage
Part Time Teacher 28 6.1 Full Time Teacher 275 60.0 Head Teacher 100 21.8 Deputy Principal 15 3.3 Leading Teacher 3 0.7 Other (Counsellor, etc) 14 3.1 Missing 23 5.0 Total 458 100.0
Table 5.2.5
Breakdown of Participating Teachers’ Years of Teaching Experience
Experience Number Approximate Percentage
1 to 2 Years 13 2.8 3 to 5 Years 39 8.5 6 to 10 Years 45 9.8 11 to 20 Years 148 32.3 20 + Years 192 41.9 Missing 21 4.6 Total 458 100.0
398
Table 5.2.6
Participating Teachers’ Length of Service at Current School
Experience Number Approximate Percentage
1 to 2 Years 69 15.1 3 to 5 Years 108 23.6 6 to 10 Years 111 24.2 11 to 20 Years 114 24.9 20 + Years 35 7.6 Missing 21 4.6 Total 458 100.0
Table 5.2.7
Participating Teachers’ Age
Teacher’s Age Number Approximate Percentage
Less than 30 Years 59 12.9 30 to 39 Years 82 17.9 40 to 49 Years 210 45.9 50 to 59 Years 79 17.2 More than 60 Years 4 0.9 Missing 24 5.2 Total 458 100.0
Table 5.2.8
Participating Teachers’ Gender
Teacher’s Gender Number Approximate Percentage
Male 200 43.7 Female 235 51.3 Missing 23 5.0 Total 458 100.0 .
399
Table 5.2.9
Participating Teachers’ Years of Service with Current Principal
Years with Current Principal
Number Approximate Percentage
3 Months or less 59 5.5 Between 3 and 6 Months 10 2.2 Between 6 and 12 Months 12 2.6 Between 1 and 2 Years 99 21.6 Over 2 Years 291 63.5 Missing 21 4.6 Total 458 100.0 Table 5.2.10
Breakdown of Participating Principals’ Age
Principal’s Age Number Approximate Percentage
Less than 30 Years 0 0.0 30 to 39 Years 0 0.0 40 to 49 Years 23 44.2 50 to 59 Years 25 48.1 More than 60 Years 1 1.9 Missing 3 5.8 Total 52 100.0
Table 5.2.11
Breakdown of Participating Principals’ Gender
Principal’s Gender Number Approximate Percentage
Male 39 75.0 Female 10 19.3 Missing 3 5.7 Total 52 100.0
400
Table 5.2.12
Breakdown of Participating Principals’ Total Years of Teaching Experience
Principal’s Teaching Experience
Number Approximate Percentage
1 to 2 Years 0 0.0 3 to 5 Years 0 0.0 6 to 10 Years 0 0.0 11 to 20 Years 2 3.8 20 + Years 47 90.4 Missing 3 5.8 Total 52 100.0
Table 5.2.13
Breakdown of Participating Principals’ Total Years of Service as a Principal
Principal’s Years as a Principal
Number Approximate Percentage
1 to 4 Years 19 36.5 4 to 6 Years 6 11.5 6 to 8 Years 11 21.2 8 to 10 Years 10 19.2 10 to 12 Years 2 3.8 12 to 14 Years 1 1.9 14 to 16 Years 0 0.0 More than 16 Years 0 0.0 Missing 3 5.8 Total 52 100.0
Table 5.2.14
Breakdown of Participating Principals’ Years of Service as a Principal in their Current
School
Principal’s Years at Present School
Number Approximate Percentage
1 to 2 Years 12 23.1 3 to 5 Years 14 26.9 6 to 10 Years 21 40.4 11 to 20 Years 2 3.8 20 + Years 0 0.0 Missing 3 5.8 Total 52 100.0
401
Table 5.2.15
Breakdown of Participating Principals’ Qualifications
Highest Qualification Number Approximate Percentage
Undergraduate Degree 13 25.0 Graduate Certificate 1 1.9 Graduate Diploma 10 19.2 Masters Degree 23 44.2 Ed.D. Degree 1 1.9 Ph.D. Degree 0 0.0 Other 1 1.9 Missing 3 5.8 Total 52 100.0 Table 5.2.16
Breakdown of Further Study Being Undertaken By Participating Principals’
Principal’s Current Area of Study
Number Approximate Percentage
Undergraduate Degree 0 0.0 Graduate Certificate 0 0.0 Graduate Diploma 0 0.0 Masters Degree 4 7.7 Ed.D. Degree 1 1.9 Ph.D. Degree 2 3.8 Other 1 1.9 Missing 44 84.6 Total 52 100.0
402
REFERENCES
306
Abolghasemi, M., McCormick, J., & Conners, R. (1999). The importance of department heads in the development of teacher support for school vision. International Journal for Educational Management, 13(2), 80-86.
Adair, J. (1984). The skills of leadership. Aldershot, United Kingdom: Gower
Publishing. Anderson, C. S. (1982). The search for school climate: A review of the research.
Review of Educational Research, 52(3), 368-420. Australian Bureau of Statistics. (2005). Schools: Australia (4221.0). Retrieved
October 31, 2005 from Australian Bureau of Statistics, Canberra. Web site: www.ausstats.abs.gov/Ausstats/subscriber.netLookup/07B7F5D9A11E44E6CA256FB1007911F4/$File/42210_2004.pdf
Avolio, B. J., Bass, B. M., & Jung, D. I. (1995). MLQ multifactor leadership questionnaire: Technical report. Palo Alto, CA: Mind Garden.
Barker, A. M. (1990). Transformational nursing leadership: A vision for the
future. Baltimore: Williams and Wilkins. Barth, R. S. (1990). Improving schools from within: Teachers, parents and
principals can make a difference. San Francisco: Jossey-Bass Inc., Publishers.
Bass, B. M. (1985a). Leadership and performance beyond expectation. New York: The Free Press.
Bass, B. M. (1985b). Leadership: Good, better, best. Organisational Dynamics, 13, 26-40.
Bass, B. M. (1988). The inspirational processes of leadership. Journal of Management Development, 7(5), 20-31.
Bass, B. M. (1990). Bass and Stogdill’s handbook of leadership: A survey of theory and research. New York: Free Press.
Bass, B. M., & Avolio, B. J. (1994). Improving organisational effectiveness
through transformational leadership. Thousand Oaks, CA: Sage Publications.
Bass, B. M., & Avolio, B. J. (1995). Transformational leadership development:
Manual for the multifactor leadership questionnaire. Palo Alto, CA: Consulting Psychologists Press, Inc.
Bass, B. M., & Avolio, B. J. (1997). The full range leadership development
manual for the multifactor leadership questionnaire. Redwood City, CA: Mindgarden Inc.
307
Bass, B. M., & Hater, J. J. (1988). Superiors’ evaluation and subordinates’ perceptions of transformational and transactional leadership. Journal of Applied Psychology, 73(4), 695-702.
Bass, B. M., Waldman, D. A., Avolio, B. J., & Bebb, M. (1987). Transformational leadership and the falling dominoes effect. Group and Organisational Studies, 12(1), 73-87.
Beare, H., Caldwell, B. J., & Millikan, R. H. (1989). Creating an excellent school. London: Billings and Sons Ltd.
Beckner, W. (1990). The why and how: Commitment and leadership. NASSP
Bulletin, 74, 4-10. Bennis, W. G., & Nanus, B. (1985). Leaders: The strategies for taking charge.
New York: Harper and Row, Publishers. Berson, Y., Shamir, B., Avolio, B. J., & Popper, M. (2001). The relationship
between vision strength, leadership style, and context. The Leadership Quarterly, 12, 53-73.
Blake, R. R., & McCanse, A. A. (1991). Leadership dilemmas—Grid solutions.
Houston, TX: Gulf Publishing Company. Blasé, J., & Blasé, J. (2000). Effective instructional leadership. Journal of
Educational Administration, 38, 130-141. Blasé, J., & Kirby, P. C. (1992). The power of praise. NASSP Bulletin, 76, 69
-77. Blasé, J., Dedrick, C., & Strathe, M. (1986). Leadership behaviour of school
principals in relation to teacher stress, satisfaction and performance. Journal of Humanistic Education and Development, 24(4), 159-171.
Bogdan, R., & Biklen, S. (1998). Qualitative research in education. Boston:
Allyn and Bacon. Bogdan, R., & Taylor, S. (1998). Introduction to qualitative research methods.
New York: John Wiley. Bogler, R. (2001). The influence of leadership style on teacher job satisfaction.
Educational Administration Quarterly, 37, 662-683. Bollen, K. A. (1989). Structural equations with latent variables. New York:
John Wiley. Brady, L. (1988). The principal as a climate factor in Australian schools: A
review of studies. Journal of Educational Administration, 26(1), 73-96.
308
Brookover, W. B., Schweitzer, J. H., Schneider, J. M., Beady, C. H., Flood, P. K., & Wisenbaker, J. M. (1978). Elementary school social climate and school achievement. American Educational Research Journal, 15, 301-318.
Browne, M. W., & Cudeck, R. (1993). Alternative ways of assessing model fit.
In K. A. Bollen, & J. S. Long (Eds.), Testing structural equation models (pp. 136-161). Newbury Park, CA: Sage Publications.
Burns, J. M. (1978). Leadership. New York: Harper and Row.
Butler, E. D. (1995). Improving school learning environments: A resource
manual of knowledge and strategies. Memphis, TN: Memphis University, Center for Research in Educational Policy.
Byrne, B. M. (1998). Structural equation modeling with LISREL, PRELIS and
SIMPLIS. Hillsdale, NJ: Erlbaum. Caldwall, B. J. (1998). Strategic leadership, resource management and school
reform. Journal of Educational Administration, 36, 445-461. Cameron, K. S. (1978). Measuring organisational effectiveness in institutions of
higher education. Administration Science Quarterly, 23, 604-632. Caracelli, V. J., & Greene, J. C. (1997). Crafting mixed-method evaluation
designs. New Directions for Evaluation, 74, 19-32. Chui, H. S., Sharpe, F. G., & McCormick, J. (1996). Vision and leadership of
principals in Hong Kong. Journal of Educational Administration, 34(3), 30-48.
Cliff, N. (1983). Some cautions concerning the application of causal modelling
methods. Multivariate Behavioural Research, 18, 115-126. Cohen, L., & Manion, L. (1984). Research methods in education (2nd ed.).
Dover, New Hampshire: Croom Helm.
Conger, J. A., & Kanungo, R. N. (1988). Charismatic leadership: The elusive factor in organisational effectiveness. San Francisco: Jossey-Bass Publishers.
Conley, S., & Levinson, R. (1993). Teacher work redesign and job satisfaction.
Educational Administration Quarterly, 29(4), 453-478. Conley, S., & Muncey, D.E. (1999). Organisational climate and teacher
professionalism: Identifying teacher work environment dimensions. In H. J. Freiberg (Ed.), School climate measuring, improving and sustaining healthy learning environments (pp. 103-123). London: Falmer Press.
309
Cotton, K. (1996). School size, school climate, and student performance. School Improvement Research Series, May, 1996. Retrieved November 20, 2002 from Northwest Regional Educational Laboratory. Web site: www.nwrel.org/scpd/sirs/10/c020.html
Creswell, J. W. (1994). Research design: Qualitative & quantitative approaches.
Thousand Oaks, CA: Sage Publications. Davies, B. (2002). Rethinking schools and school leadership for the twenty-first
century: Changes and challenges. International Journal of Educational Management, 16, 196-206.
Davis, J., & Wilson, S. M. (2000). Principals’ efforts to empower teachers:
Effects on teacher motivation and job satisfaction and stress. The Clearing House, 73, 349-353.
Day, C. (2000). Beyond transformational leadership. Educational Leadership,
57, 56-60. De Roche, E. F. (1985). How school administrators solve problems. New York:
Prentice-Hall, Inc. Densten, I., & Sarros, J. (1997, July). Re-thinking transformational leadership.
Working Paper (59/97). Monash University, Victoria, Australia: Faculty of Business and Economics.
Dereshiwsky, M. I. (1999). Qualitative research. Retrieved May 15, 2002 from
http://jan.ucc.nau.edu/~mid/edr725/class/ Dinham, S., & Scott., C. (1997). Moving into the third, outer domain of teacher
satisfaction. Unpublished manuscript, University of Western Sydney, NSW, Australia.
Dinham, S., & Scott., C. (1998). A three domain model of teacher and school
executive career satisfaction. Journal of Educational Administration, 36, 362-378.
Dinham, S., Cairney, T., Craigie, D., & Wilson, S. (1995). School climate and
leadership: Research into three secondary schools. Journal of Educational Administration, 33, 36-58.
Dionne, S. D., Yammarino, F. J., Atwater, L.E., & James, L.R. (2002).
Neutralising substitutes for leadership theory: Leadership effects and common-source bias. Journal of Applied Psychology, 87, 545-464.
Docker, J. G., Fraser, B. J., & Fisher, D. L. (1989). Differences in the
psychosocial work environment of different types of schools. Journal of Research in Childhood Education, 4(1), 5-17.
310
Dorman, J. P., Fraser, B. J., & McRobbie, C. J. (1995). Relationship between school-level and classroom-level environments in secondary schools. Journal of Educational Administration, 35(1), 74-91.
du Toit, S., du Toit, M., & Cudeck, R. (1999). Introduction to the analysis of
multilevel models with Lisrel 8.30. Chicago: Scientific Software International.
Duignan, P. A. (1986). Research on effective schooling: Some implications for
school improvement. Journal of Educational Administration, 24(1), 59-73. Duignan, P. A. (1989). Reflective management: The key to quality leadership. In
C. R. Riches, & C. Morgan (Eds.), Human resource management in education (pp. 74-90). Philadelphia: The Open University Press.
Duke, D. L. (1998). Challenges of designing the next generation of America’s
schools. Phi Delta Kappan, 79, 688-693. Eicholtz, R. L. (1984). School climate: Key to excellence. American Education,
20(1), 22-26. Elliott, C. (1992). Leadership and change in schools. Issues in Educational
Research, 2, 45-55. English, F. W. (1992). The principal and the prince: Machiavelli and school
leadership. NASSP Bulletin, 76(540), 10-17. Fisher, D. L., & Fraser, B. J. (1990, April). Validity and use of the school-level
environment questionnaire. Paper presented at the annual meeting of the American Educational Research Association, Boston, MA.
Fisher, D. L., & Fraser, B. J. (1991). Validity and use of school environment
instruments. Journal of Classroom Interaction, 26(2), 13-18. Fraser, B. J. (1983). Use of classroom environment instruments in person
-environment fit research. Perth, Australia: Western Australian Institute of Technology.
Fraser, B. J. (1986). Classroom environment. London: Croom Helm.
Fraser, B. J. (1999). Using learning environment assessments to improve
classroom and school climates. In H.J. Freiberg (Ed.), School climate measuring, improving and sustaining healthy learning environments (pp. 65-83). London: Falmer Press.
Fraser, B. J., & Rentoul, A. J. (1982). Relationships between school-level and
classroom-level environment. Alberta Journal of Educational Research, 28(3), 212-225.
311
Fraser, B. J., Anderson, G. J., & Walberg, H. J. (1982). Assessment of learning environments: Manual for Learning Environment Inventory (LEI) and My Class Inventory (MCI). Perth, Australia: Western Australian Institute of Technology.
Fraser, B. J., Williamson, J. C., & Tobin, K. G. (1987). Use of classroom and school climate scales in evaluating alternative high schools. Teaching and Teacher Education, 3(3), 219-231.
Freiberg, H. J. (1998). Measuring school climate: Let me count the ways.
Educational Leadership, 56(1), 22-27.
Freiberg, H. J., & Stein, T. A. (1999). Measuring, improving and sustaining healthy learning environments. In H.J. Freiberg (Ed.), School climate measuring, improving and sustaining healthy learning environments (pp. 11-29). London: Falmer Press.
Fullen, M. G., & Hargreaves, A. (1991). What’s worth fighting for? Working
together for your school.Toronto, Canada: Ontario Public School Teachers Federation.
Gay, L. R. (1992). Educational research: Competencies for analysis and
application (4th ed.). New York: Macmillan Publishing Company. Geijsel, F., Sleegers, P., Leithwood, K. A., & Jantzi, D. (2002).
Transformational leadership effects on teachers’ commitment and effort towards school reform. Journal of Educational Administration, 41, 228-256.
Goldstein, H. (1995). Multilevel statistical models. London: Edward Arnold. Goldstein, H., Rasbash, J., Plewis, I., Draper, D., Browne, W., Yang, M.,
Woodhouse, G. & Healy, M. (1998). A user’s guide to MLwiN. London: Multilevel Models Project, Institute of Education, University of London.
Grbich, C. F. (1999). Qualitative research in health: An introduction. St.
Leonards, Sydney, Australia: Allen and Unwin. Griffith, J. (1999). The school leadership/school climate relation: Identification
of school configurations associated with change in principals. Educational Administration Quarterly, 35(2), 267-291.
Hall, G. E. (1987, April). The principal’s role in setting school climate (for
school improvement). Paper presented at annual meeting of the American Educational Research Association, Washington, DC.
Hall, G. E., & George, A. A. (1999). The impact of principal change facilitator
style on school and classroom culture. In H. J. Freiberg (Ed.), School climate measuring, improving and sustaining healthy learning environments (pp. 165-185). London: Falmer Press.
312
Hallinger, P., & Heck, R. H. (1998). Exploring the principal’s contribution to school effectiveness: 1980-1995. School Effectiveness and School Improvement, 9, 157-191.
Halpin, A. W., & Croft, D. (1963). The organisational climate of schools.
Chicago: Midwest Administrative Center, University of Chicago. Harris, P. R. (1989). High performance leadership: Strategies for maximum
career productivity. Glenview, Ill: Scott, Foresman and Company. Hill, P., & Rowe, K. (1996). Multilevel modelling in schools effectiveness
research. School Effectiveness and School Improvement, 7, 1-34. Hollander, E. P. (1978). Leadership dynamics: A practical guide to effective
relationships. New York: The Free Press. Hoover, N. R., Petrosko, J. M., & Schulz, R. R. (1991, April). Transformational
and transactional leadership: An empirical test of a theory. Paper presented at the annual meeting of the American Educational Research Association, Chicago, Ill.
House, R. J. (1971). A path goal theory of leader effectiveness. Administrative
Science Quarterly, 16, 321-329. House, R. J., & Baetz, M. L. (1990). Leadership: Some empirical generalisations
and new research directions. In L.L. Cummings, & B.M. Staw (Eds.), Leadership, participation, and group behaviour (pp. 1-84). Greenwich, CT: JAI Press Inc.
Hoy, W. K., & Clover, S. I. R. (1986). Elementary school climate: A revision of
the OCDQ. Educational Administration Quarterly, 22(1), 93-110. Hoy, W. K., & Feldman, J. A. (1999). Organisational health profiles for high
schools. In H. J. Freiberg (Ed.), School climate measuring, improving and sustaining healthy learning environments (pp. 84-102). London, Great Britain: Falmer Press.
Hoy, W. K., & Ferguson, J. (1989). A theoretical framework and exploration of
organisational effectiveness of schools. In J. L. Burdin (Ed.), School leadership: A contemporary reader (pp. 259-274). Newbury Park, CA: Sage Publications.
Hoy, W. K., & Miskel, C. G. (1982). Educational administration: Theory,
research, and practice (2nd ed.). New York: Random House. Hoy, W. K., & Miskel, C. G. (1987). Educational administration: Theory,
research, and practice (3rd ed.). New York: Random House. Hoy, W. K., & Miskel, C. G. (1996). Educational administration: Theory,
research, and practice (5th ed.). New York: McGraw-Hill, Inc.
313
Hoy, W. K., Tarter, C. J., & Bliss, J. R. (1990). Organisational climate, school health, and effectiveness: A comparative analysis. Educational Administration Quarterly, 26(3), 260-279.
Ingram, P. D. (1997). Leadership behaviours of principals in inclusive educational settings. Journal of Educational Administration, 35, 411-427.
James, L. R., & Jones, A. P. (1974). Organisational climate: A review of theory
and research. Psychological Bulletin, 81, 1096-1112. Johnson, J. P., Livingston, M., Schwartz, R. A., & Slate, J. R. (2000). What
makes a good elementary school? A critical examination. Journal of Educational Research, 93, 339-353.
Jöreskog, K. G. & Sörbom, D. (1993). Lisrel 8: Users reference guide. Chicago:
Scientific Software International, Inc. Jöreskog, K. G. & Sörbom, D. (1996). Lisrel 8: Users reference guide. Chicago:
Scientific Software International, Inc. Jöreskog, K. G. & Sörbom, D. (2003). Lisrel (Version 8.54) [Computer
Software]. Chicago: Scientific Software International, Inc. Keefe, J. W., & Kelley, E. A. (1990). Comprehensive assessment and school
improvement. NASSP Bulletin, 74(530), 54-63. Keefe, J. W., Kelley, E. A., & Miller, S. K. (1985). School climate: Clear
definitions and a model for a larger setting. NASSP Bulletin, 69(484), 70-77.
Keeves, P. J., & Sellin, N. (1988). Multilevel analysis. In J. P. Keeves (Ed.),
Educational research, methodology and measurement: An international handbook (pp. 689-700). Oxford, United Kingdom: Pergamon Press.
Kellerman, B. (1984). Leadership: Multidisciplinary perspectives. Englewood
Cliffs, NJ: Prentice-Hall, Inc. Kelley, E. A. (1989). Improving school climate. NASSP Bulletin, 15(4), 1-6. Kelloway, E. K. (1998). Using Lisrel for structural equation modelling: A
researcher’s guide. Thousand Oaks, CA: Sage Publications.
Kerr, S., & Jermier, J. M. (1978). Substitutes for leadership: Their meaning and measurement. Organisational Behaviour and Human Performance, 22, 375-403.
Kerr, S., Schriesheim, C. A., Murphy, C. J. & Stogdill, R. M. (1974). Toward a
contingency theory of leadership based upon the consideration and initiating structure literature. Organisational Behaviour and Human Performance, 12, 62-82.
314
Kirby, P. C., King, M. I., & Paradise, L. V. (1992). Extraordinary leaders in education: Understanding transformational leadership. Journal of Educational Research, 85(5), 303-311.
Koene, B. A. S., Vogelaar, A. L. W., & Soeters, J. L. (2002). Leadership effects on organisational climate and financial performance: Local leadership effect in chain organisations. The Leadership Quarterly, 13, 193-215.
Kotter, J. P. (1988). The leadership factor. New York: The Free Press.
Kottkamp, K. B., Mulhern, J. A., & Hoy, W. K. (1987). Secondary school
climate: A revision of the OCDQ. Educational Administration Quarterly, 23(3), 31-48.
Lake, S. (1991). Defining an effective climate for a middle level school
(Practitioner’s Monograph No. 11). Irvine, CA: California League of Middle Schools.
Lambert, P. (1990). Establishing an open climate for professional development
(Practically Speaking No. 5). Sydney, Australia: New South Wales Department of Education.
Lashway, L. (2000). Leading with vision. Eugene, OR: ERIC Clearinghouse on
Educational Management. Latham, A. S. (1998). Teacher satisfaction. Educational Leadership, 55(5), 82-
84. Lee, V. E., Dedrick, R. F., & Smith, J. B. (1991). The effect of the social
organisation on teachers’ efficacy and satisfaction. Sociology of Education, 64, 190-208.
Leithwood, K. A. (1992). The move towards transformational leadership.
Educational Leadership, 49, 8-12. Leithwood, K. A., & Jantzi, D. (1990, June). Transformational leadership: How
principals can help reform school cultures. Paper presented at the annual meeting of the Canadian Association for Curriculum Studies, Victoria, BC.
Leithwood, K. A., & Jantzi, D. (1997). Explaining variations in teachers’
perceptions of principals’ leadership: A replication. Journal of Educational Administration, 35, 312-331.
Leithwood, K. A., & Jantzi, D. (2000). The effects of transformational
leadership on organisational conditions and student engagement with schools. Journal of Educational Administration, 38, 112-129.
Leithwood, K. A., & Steinbach, R. (1991). Indicators of transformational
leadership in the everyday problem solving of school administrators. Journal of Personnel Evaluation in Education, 4(3), 221-244.
315
Leithwood, K. A., Begley, P. T., & Cousins, J. B. (1990). The nature, causes and consequences of principals’ practices: An agenda for future research. Journal of Educational Administration, 28, 5-31.
Leithwood, K. A., Jantzi, D., & Steinbach, R. (1999). Changing leadership for changing times. Buckingham, Philadelphia: Open University Press.
Leithwood, K. A., Jantzi, D., Silins, H., & Dart, B. (1992, January).
Transformational leadership and school restructuring. Paper presented at the International Congress for School Effectiveness and Improvement, Victoria, BC.
Lezotte, L. W. (1991). Correlates of school effectiveness: The first and second
generation. Okemos, MI: Effective Schools Products, Ltd. Licata, J.W., & Harper, G.W. (2001). Organisational health and robust school
vision. Educational Administration Quarterly, 37, 5-26. Lichtman, M. (2001). Data analysis in qualitative research. Retrieved July 6,
2002, from Virginia Polytechnic Institute and State University. Web Site: http://www.nvgc.vt.edu/mlichtman/qual2001/Analysis.doc
Likert, R., & Likert, J. G. (1976). New ways of managing conflict. New York:
McGraw-Hill, Inc. Lindelow, J., Mazzarella, J. A., Scott, J. J., Ellis, T. I., & Smith, S. C. (1989).
School climate. In S.C. Smith, & P. K. Piele (Eds.), School leadership: Handbook for excellence (pp. 168-188). Washington, D.C.: Office of Educational Research and Improvement.
Lockwood, A. T. (1997). Conversations with educational leaders:
Contemporary viewpoints on education in America. New York: State University of New York Press.
Lunenburg, F. C., & Ornstein, A. C. (2000). Educational administration.
Concepts and practices (3rd ed.). Belmont, CA: Wadsworth. Mann, D. (1992). School reform in the United States: A national policy review
1965-91. School Effectiveness and School Improvement, 3, 216-230. Marsh, H. W. & Hau, K. T. (2004). Explaining paradoxical relations between
academic self-concepts and achievements: Cross-cultural generalizability of the internal-external frame of reference predictions across 26 countries. Journal of Educational Psychology, 96, 56-67.
Marsh, H. W., Balla, J. R., & Hau, K. T. (1996). An evaluation of incremental
fit indices: A clarification of mathematical and empirical processes. In G. A. Marcoulides, & R. E. Schumacker (Eds.), Advanced structural equation modeling techniques (pp. 315-353). Hillsdale, NJ: Erlbaum.
316
Marsh, H. W., Balla, J. R., & McDonald, R. P. (1988). Goodness-of-fit indices in confirmatory factor analysis: The effect of sample size. Psychological Bulletin, 102, 391-410.
Marsh, H. W., Kong, C. K., & Hau, K. T. (1999). Longitudinal multilevel models of
the big fish little pond effect on academic self-concept: Counterbalancing contrast and reflected glory effects in Hong Kong schools. Canberra, Australia: Australian Research Council. (Eric Document Reproduction Service No: ED442790)
Marshall, G., & Rossman, G. (1999). Designing qualitative research. Thousand
Oaks, CA: Sage Publications. Maxwell, T. W., & Thomas, A. R. (1991). School climate and school culture.
Journal of Educational Administration, 29(2), 72-82. Midgley, C., Maehr, M., Hicks, L., Roeser, R., Urdan, T., Anderman, E., Kaplan, A.
(1996). Patterns of Adaptive Learning Survey. Ann Arbor, MI: University of Michigan.
Miles, M. B., & Huberman, A. M. (1994). Qualitative data analysis: An expanded
sourcebook (2nd ed.). Thousand Oaks, CA: Sage Publications. Minichiello, V., Aroni, R., Timewell, E., & Alexander, L. (1995). In-depth
interviewing: Principles, techniques, analysis (2nd ed.). Melbourne, Australia: Longman.
Miskel, C., & Ogawa, R. (1988). Work motivation, job satisfaction, and climate. In
N. J. Boyan (Ed.), Handbook of research on educational administration (pp. 279-304). New York: Longman.
Mok, M., & Flynn, M. (1997). Does school size affect quality of school life? Issues
in Educational Research, 7(1), 69-86. Murray, F., & Feitler, F. C. (1989, March). An investigation of transformational
leadership and organisational effectiveness in small college settings. Paper presented at the annual meeting of American Educational Research Association, San Francisco, CA.
New South Wales Department of Education and Training. (2000). Leading and
managing the school: A statement of key accountabilities for principals in the effective educational leadership and management of NSW government schools [Electronic version]. Retrieved October 13, 2004, from <https://www.det.nsw.edu.au/policies/general_man/accountability/lead_sch/pd04_24_manage_lead.pdf>.
Northouse, P. G. (2004). Leadership: Theory and practice (3rd ed.). Thousand
Oaks, CA: Sage Publications.
317
Norusis, M. J. (1993). SPSS base system user’s guide. New York: McGraw Hill Book Company/SPSS Inc.
Nunnally, J. (1978). Psychometric theory. New York: McGraw Hill.
Owens, R. G. (1987). Organisational behaviour in education (3rd ed.). Englewood
Cliffs, NJ: Prentice-Hall, Inc. Owens, R. G. (1998). Organisational behaviour in education (6th ed.). Boston:
Allyn and Bacon. Papaioannou, A., Marsh, H. W., & Theodorakis, Y. (2003). A multilevel approach
to motivational climate in physical education and sports settings: An individual or group level construct? Manuscript submitted for publication.
Parry, K. W., & Sarros, J. C. (1996). An Australian perspective on transformational
leadership. In K. W. Parry (Ed.), Leadership research and practice: Emerging themes and new challenges (pp. 105-112). South Melbourne, Australia: Pitman Publishing.
Pasi, R. J. (2001). A climate for achievement. Principal Leadership, 2(4), 17-20. Patton, M. Q. (1990). Qualitative evaluation and research methods. Newbury Park,
CA: Sage Publications. Purkey, C. S., & Smith, M. S. (1983). Effective schools: A review. The Elementary
School Journal, 83(4), 427-452. Rasbash, J., Browne, W., Goldstein, H., Yang, M., Plewis, I., Healy, M.,
Woodhouse, G., Draper, D., Langford, I., Lewis, T. (2000). A user’s guide to MLWiN. London: University of London, Centre of Multilevel Modelling, Institute of Education.
Rentoul, A. J., & Fraser, B. J. (1983). Development of a school-level environment
questionnaire. Journal of Educational Administration, 21(1), 21-39. Robbins, S. P., Millett, B., Cacioppe, R., & Waters-Marsh, T. (1998).
Organisational behaviour. Leading and managing in Australia and New Zealand (2nd ed.). Sydney: Prentice Hall.
Roueche, J. E., Baker, G. A., & Rose, R. R. (1989). Shared vision:
Transformational leadership in American community colleges. Washington, DC: The Community College Press.
Rowe, K. J. (2000a). Multilevel analysis with MLn/MLWin and Lisrel 8.30: An
integrated course. The 16th ACSPRI Summer Program in Social Research Methods and Research Technology, The Australian National University. Centre for Applied Educational Research, University of Melbourne.
318
Rowe, K. J. (1995, February). Multilevel analysis: ACSPRI-95 course notes. Canberra, Australia: The Australian National University.
Rowe, K. J. (2000b). Simultaneous estimation of interdependent effects among
multilevel composite variables in psychosocial research: An annotated example of the application of multilevel structural equation modelling. In N. Duan, & S. Reise (Eds.), Multilevel modelling: Methodological advances, issues and applications (pp. 1-37). Hillsdale, NJ: Erlbaum.
Sagor, R. D. (1991, October). Operationalising transformational leadership: The
behaviour of principals in fostering teacher centered school development. Paper presented at the University Council for Educational Administration Annual Conference, Baltimore, MA.
Saskhin, M. (1988). The visionary principal: School leadership for the next century.
Education and Urban Society, 20(3), 239-249. Schmidt, G. L. (1976). Job satisfaction among secondary school administrators.
Educational Administration Quarterly, 12, 81. Schwartz, H., & Jacobs, J. (1979). Qualitative sociology: A method to the madness.
New York: Free Press. Sergiovanni, T. J. (1967). Factors which affect satisfaction and dissatisfaction of
teachers. Journal of Educational Administration, 5, 66-82. Sergiovanni, T. J. (1990a). Value-added leadership: How to get extraordinary
performance in schools. San Diego, CA: Harcourt Brace Jovanovich. Sergiovanni, T. J. (1990b). Adding value to leadership gets extraordinary results.
Educational Leadership, 47(8), 23-27. Sergiovanni, T. J. (1991). The principalship: A reflective practice perspective.
Boston: Allyn and Bacon. Sergiovanni, T. J., & Carver, F. D. (1980). The new school executive: A theory of
administration (2nd ed.). New York: Harper and Row, Publishers. Short, P. M., & Greer, J. T. (2002). Leadership in empowered schools: Themes from
innovative efforts. Upper Saddle River, NJ: Merrill Prentice Hall. Shum, L. C., & Cheng, Y. C. (1997). Perceptions of women principals’ leadership
and teachers’ work attitudes. Journal of Educational Administration, 35, 165-184.
Silins, H. C. (1994). The relationship between transformational leadership and
school improvement outcomes. School Effectiveness and School Improvement, 5, 272-298.
319
Silins, H. C., & Murray-Harvey, R. (1999). What makes a good senior secondary school? Journal of Educational Administration, 37, 329-344.
Silver, P. F. (1983). Educational administration: Theoretical perspectives on
practice and research. New York: Harper and Row, Publishers. Spence, K. (1991). School leadership—The role of the principal: Towards a criteria
framework. Independent Education, 21, 14-17. Stanton, M. J. (1999). Schools that teach: A blueprint for the millennium. US
Today (Magazine), July 01. Retrieved May 28, 2002 from High Beam Research. Web site: http://static.highbeam.com/u/usatodaymagazine/july011999/schoolsthatteachablueprintforthemillenniumupdating/
Starrett, R. J. (1991). Building an ethical school: A theory for practice in
educational leadership. Educational Administration Quarterly, 27 (2), 185-202.
Starrett, R. J. (1993). Transforming life in schools: Conversations about leadership
and school renewal. Hawthorn, Victoria: Australian Council for Educational Administration.
Starrett, R. J. (2004a). Ethical Leadership. San Francisco, CA: Jossey-Bass. Starrett, R. J. (2004b). The dialogue of scholarship. Educational Administration
Quarterly, 40 (2), 259-267. Sweeney, J. (1982). Research on effective school leadership. Educational
Leadership, 39(5), 346-351. Swenson, D. X. (2000). Dysfunctional leaders. Dealing with flawed leadership:
Substitutes, Neutralizers & Enhancements. Retrieved August 4, 2003, from The College of St. Scholastica, Duluth, MN. Web Site: http://www.ccs.edu/users.dswenson/web/LEAD/substitutes.html
Tuckman, B. W. (1988). Conducting Educational Research (3rd ed.). San Diego,
CA: Harcourt Brace Jovanovich. Van Fleet, D. D., & Yukl, G. A. (1989). A century of leadership research. In W. E.
Rosenbach, & R. L. Taylor (Eds.), Contemporary issues in leadership (pp. 65-90). San Francisco: Westview Press.
Vandenberghe, R. & Staessens, K. (1991, April). Vision as a core component in
school culture. Paper presented at the annual meeting of the American Educational Research Association, Chicago, Ill.
Vandenberghe, R. (1995). Creative management of schools: A matter of vision and
daily interventions. Journal of Educational Administration, 33, 31-51.
320
Walker, J. E. (1990). The skills of exemplary principals. NASSP Bulletin, 74(524), 48-55.
Weick, K. (1976). Educational organisations as loosely coupled systems.
Administrative Science Quarterly, 21, 1-9. Wolcott, H. (1982). Differing styles of on-site research, or, if it isn’t ethnography
what is it? The Review Journal of Philosophy and Social Science, 7, 154-169. Worthen, B. R., Sanders, J. R., & Fitzpatrick, J. L. (2004). Program evaluation:
Alternative approaches & practical guidelines (3rd ed.). Boston: Allyn and Bacon.
Wyatt, T. (1996). School effectiveness research: Dead end, damp squib or
smouldering fuse? Issues in Educational Research, 6, 79-112. Young, D. J. (1998, April). Characteristics of effective rural schools: A
longitudinal study of Western Australian rural high school students. Paper presented to the annual meeting of the American Educational Research Association, San Diego, CA.
Young, D. J., & Fraser, B. J. (1992, April). Sex differences in science achievement:
A multilevel analysis. Paper presented at the annual meeting of the American Educational Research Association, San Francisco, CA.
Yukl, G. A. (1971). Toward a behavioural theory of leadership. Organisational
Behaviour and Human Performance, 6, 414-440. Yukl, G. A. (2002). Leadership in organisations (5th ed.). Upper Saddle River, NJ:
Prentice Hall.
321