This article was downloaded by: [Universite Laval]On: 18 July 2012, At: 08:55Publisher: RoutledgeInforma Ltd Registered in England and Wales Registered Number: 1072954 Registeredoffice: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

The International Journal of HumanResource ManagementPublication details, including instructions for authors andsubscription information:http://www.tandfonline.com/loi/rijh20

Organizational citizenship behaviourand employee retention: how importantare turnover cognitions?Pascal Paillé aa Departement Management, Université Laval, Québec, Canada

Version of record first published: 17 Jul 2012

To cite this article: Pascal Paillé (2012): Organizational citizenship behaviour and employeeretention: how important are turnover cognitions?, The International Journal of Human ResourceManagement, DOI:10.1080/09585192.2012.697477

To link to this article: http://dx.doi.org/10.1080/09585192.2012.697477

PLEASE SCROLL DOWN FOR ARTICLE

Full terms and conditions of use: http://www.tandfonline.com/page/terms-and-conditions

This article may be used for research, teaching, and private study purposes. Anysubstantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,systematic supply, or distribution in any form to anyone is expressly forbidden.

The publisher does not give any warranty express or implied or make any representationthat the contents will be complete or accurate or up to date. The accuracy of anyinstructions, formulae, and drug doses should be independently verified with primarysources. The publisher shall not be liable for any loss, actions, claims, proceedings,demand, or costs or damages whatsoever or howsoever caused arising directly orindirectly in connection with or arising out of the use of this material.

Organizational citizenship behaviour and employee retention:how important are turnover cognitions?

Pascal Paille*

Departement Management, Universite Laval, Quebec, Canada

The purpose of this study is to examine the relationships between perceived jobalternatives, intention to search, intention to leave and organizational citizenshipbehaviour (OCB), with a view to determining the extent to which these variables arerelated. Two surveys using separate samples (Study 1, n ¼ 651, and Study 2, n ¼ 226)were used. First, in both studies, perceived job alternatives provided a better account ofOCB towards the organization (sportsmanship and civic virtue) than OCB towardsindividuals (helping and altruism). Second, the results of Studies 1 and 2 indicate thatthe relationships between OCB and intention to search and between OCB and intentionto quit are different. The findings suggest that one part of the research model appears tobe generalizable, while the other part appears to be explained by the context ofemployment. The implications of the findings are discussed.

Keywords: intention to leave; intention to search a job; organizational citizenshipbehaviors; perceived job alternatives

Introduction

Although there is evidence demonstrating the influence of organizational citizenship

behaviour (OCB) on employee retention (Chen, Hui and Sego 1998; MacKenzie, Podsakoff

and Ahearne 1998; Chen 2005; Harrison, Newman and Roth 2006; Coyne and Ong 2007;

Lam, Chen and Takeuchi 2009; Paille and Grima 2011), some studies have argued that

further research is needed to improve our understanding of the relationships between OCBs

and variables involved in the decision of employees to resign from their organizations

(Organ, Podsakoff and MacKenzie 2006, p. 239). The aim of this paper is to extend the chain

of relationships between OCB and employee retention by examining the influence of

variables typically involved in turnover cognitions, such as perceived job alternatives (i.e.

favourability of the external job environment), intention to search (i.e. looking for a job

outside the organization) and intention to leave (i.e. leaving the organization at some

unspecified point in the future). In so doing, the paper addresses two specific issues. First,

with the exception of a small number of studies (Coyne and Ong 2007; Paille and Grima

2011), the majority of studies linking OCB and employee retention (intention to leave and

actual turnover) use a global measure (Chen et al. 1998; Koys 2001; Chen 2005; Mossholder,

Settoon and Hanagan 2005; Krishnan and Singh 2010). The specific impact of forms of OCB

on the propensity to leave therefore remains largely under-researched. Second, with the

exception of a small number of studies (Hui, Law and Chen 1999; Thau, Bennett, Stahlberg

and Werner 2004), the relationships between OCB and dimensions of turnover cognitions

other than intention to leave the organization have been largely overlooked. The distinctions

ISSN 0958-5192 print/ISSN 1466-4399 online

q 2012 Taylor & Francis

http://dx.doi.org/10.1080/09585192.2012.697477

http://www.tandfonline.com

*Email: pascal.paille@fsa.ulaval.ca

The International Journal of Human Resource Management,

2012, 1–23, iFirst

Dow

nloa

ded

by [

Uni

vers

ite L

aval

] at

08:

55 1

8 Ju

ly 2

012

between situations in which employees perceive job alternatives reflecting their potential

value in the labour market and situations in which employees actively search for a job in

order to leave their current employer are therefore difficult to establish.

The paper examines the relationships between OCB, perceived job alternatives,

intention to search and intention to leave the organization in order to determine the

relationship between OCB and turnover cognitions. The paper is organized as follows. It

begins by providing a literature review before presenting two field studies. The paper

concludes by discussing the implications of the results for future research on the

relationships between OCB and turnover cognitions.

Literature review

Organizational citizenship behaviour and employee retention

According to Organ (1988), OCB refers to:

individual behaviour that is discretionary, not directly or explicitly recognized by the formalreward system, and that in the aggregate promotes the effective functioning of theorganization. By discretionary, we mean that the behaviour is not an enforceable requirementof the role or the job description, that is, the clearly specifiable terms of the person’semployment contract with the organization; the behaviour is rather a matter of personalchoice, such that its omission is not generally understood as punishable. (p. 4)

The use of OCB for studying employee retention is a relatively recent research interest,

even though a substantial amount of empirical research has already been published on the

topic. According to Chen (2005), this research interest is connected at least in part to the

findings outlined in two meta-analyses (Hom and Griffeth 1995; Griffeth, Hom and Gaertner

2000), indicating that the variables commonly involved in the study of the withdrawal

process (organizational commitment and job satisfaction) only account for a very small

percentage of variance (approximately 5%). Chen argued that a concept commonly

overlooked in research (OCB) enters into the retention equation. In addition to the arguments

given by Chen, a conceptual explanation may also account for this interest. OCBs are

common and widespread forms of behaviour in the work environment. Providing support to

a colleague, helping a recently recruited employee to settle into the organization,

demonstrating a degree of flexibility by tolerating requests deemed to be excessive or

unreasonable, or defending the image of the organization in a discussion are not neutral

forms of behaviour, given the frame of mind they suggest and their positive impact on the

work environment. Social exchange theory describes the foundations of this frame of mind.

Employees demonstrate civic behaviour when they feel that they are supported and treated

fairly by their employer. Though agreeing with the conceptual framework of social exchange

theory, Lavelle (2010) argued that demonstrating OCB entails personal motivations that go

beyond the need to offer something in return for being treated fairly. Whatever the

underlying reason or motivation, voluntary cooperation through OCB reflects a propensity to

go beyond the expectations of the employer and the employee’s colleagues. Voluntary

efforts, particularly those made freely and that involve going beyond the tasks explicitly

required as part of the job description, appear to be incompatible with the intention to leave.

Furthermore, the literature on social exchange theory provides findings, which indicate

that employees exchange desirable outcomes in return for fair treatment, support or care.

Since OCB enhances organizational effectiveness, top management values OCBs as

desirable outcomes. Morrison (1994) theorized the link between OCB and employee

retention, arguing that if ‘an organization’s human resource philosophy is one that places

high value on retaining employees in a long-term relationship, employees will engage in

P. Paille2

Dow

nloa

ded

by [

Uni

vers

ite L

aval

] at

08:

55 1

8 Ju

ly 2

012

more organizational citizenship behaviour’ (p. 503). The norm of reciprocity is the

underlying process providing a better account of the kind of exchange described by

Morrison. Reciprocity between two entities (e.g. an employee and an employer) requires

two basic premises: (1) people should help those who have helped them, and (2) people

should not injure those who have helped them (Gouldner 1960). On the basis of the norm

of reciprocity, Schaninger and Turnipssed (2005) argued that an employee gives

something to the donor (employer, supervisor or colleague) who provides something that

the employee finds valuable. Employees value supportive decision-making and fair

treatment, whereas employers value loyalty, in-role performance and OCB.

Organizational citizenship behaviour and the withdrawal process

Following Harrison et al. (2006), OCBs can be viewed as one of the early stages of the

withdrawal process. Before explaining the findings reported by Harrison et al. it is important

to define the nature of the withdrawal process. Following Gupta and Jenkins (1983), ‘only

those employee responses specifically designed by an individual to increase distance from

the organization can be labeled “withdrawal”’ (p. 63). On the basis of this definition, they

introduced a distinction between voluntary and involuntary behaviours to explain the kind

of individual responses linked to withdrawal. The examples given by the authors

(mandatory retirement, jury duty, an automobile accident on the way to work, tardiness to

accommodate children’s schedules or personal illness) reflect specific situations that are not

related to the employee’s willingness to increase the distance with the employer.

The withdrawal process encompasses a variety of behaviours, each reflecting a form of

voluntary response used by an employee to cope with an unpleasant work environment.

Lateness, absenteeism and turnover are typical behaviours related to the withdrawal

process (Rosse 1988; Johns 2002; Berry, Lelchook and Clark 2011). However, as

suggested by Lee, Mitchell, Sablynski, Burton and Holtom (2004), the withdrawal process

also includes behaviours such as poor citizenship and poor job performance. Chen (2005)

suggested recently that the lack of willingness to exhibit OCB (i.e. poor citizenship) may

be an indication of employee withdrawal from the organization. OCBs are discretionary in

nature and, as such, cannot be formally required by the employer or the superior (Organ

et al. 2006). Typically, if an employee demonstrates low OCB, s/he cannot be sanctioned.

Therefore, when employees experience lasting dissatisfaction with their job or become

less committed to their organization but are unable to leave because of the lack of external

job opportunities (low job alternatives), it is less risky for employees to reduce OCB than

to express discontent by reducing, for example, their efforts at work. Unlike employees

who reduce their OCB without fear of blame, employees may face sanctions if they

perform their work poorly (Tolich 1993), procrastinate (Ferrari 1992), are frequently late

(Sagie, Koslowsky and Amichai Hamburger 2002), or increase their absenteeism (Dalton

and Mesch 1991). Low OCB can be viewed as a signal that employees are beginning to

disengage from the workplace (Chen et al. 1998; Chen 2005). Harrison et al. (2006)

provided empirical support for Chen’s contention. They tested five models in order to

determine which model offers the best fit between overall job attitude and five behavioural

criteria: job performance, OCB examined as contextual performance, lateness, absence

and turnover (see their paper for further details, pp. 309–312). Although a slight

improvement was found in the data when OCB was taken into account in the withdrawal

sequence, Harrison et al. concluded that OCB can reflect ‘the first element in the

progression sequence among disturbance terms (i.e. contextual performance to lateness,

lateness to absence, and absence to turnover)’ (p. 318).

The International Journal of Human Resource Management 3

Dow

nloa

ded

by [

Uni

vers

ite L

aval

] at

08:

55 1

8 Ju

ly 2

012

Forms of OCB and turnover cognitions

When OCBs are widespread, valued and sustained within an organization, the best

employees tend to be retained (Podsakoff, MacKenzie, Paine and Bachrach 2000).

However, as noted above, most studies calculate a mean score for OCB, thus making it

impossible to identify which form of OCB is related to turnover (actual or intended). An

analysis of the relationships between OCB and employee retention based on a distinction

between OCB towards individuals and OCB towards the organization significantly

improves our understanding of this issue. OCB towards individuals implies face-to-face

actions, whereas OCB towards the organization appears to be more diffuse since its

boundaries are often difficult to perceive for individuals.

OCB towards individuals involves helping others (sometimes referred to as altruism).

Helping encompasses a set of behaviours such as courtesy, cheerleading and peacemaking.

Researchers have often conceptualized and operationalized helping as a second-order

latent factor. Helping others refers to cooperative and spontaneous behaviours that involve

providing assistance to others or preventing the occurrence of work-related problems and

contributing to interpersonal harmony (Cirka 2005). The OCB literature indicates that

employees can provide help to others in various ways, such as replacing an absent

colleague and helping another colleague to carry out a difficult task. Cheerleading can also

be viewed as a helping behaviour when an employee encourages a co-worker who is

discouraged about his accomplishment or professional development. Helping others or

receiving help from others fosters cohesiveness among employees and satisfactory

relationships with co-workers. Theoretical arguments have been developed (Podsakoff and

MacKenzie 1997; Podsakoff et al. 2000; Organ et al. 2006) to explain why and how

helping could be related to employee retention. Chen et al. (1998) point out that ‘groups

(or organizations) that have higher levels of OCB will have lower levels of turnover

because interactions among employees who exhibit high levels of OCB are likely to foster

group attractiveness and cohesiveness and subsequently to decrease voluntary turnover’

(p. 928). As argued by Organ et al. (2006), helping behaviours may enhance morale, group

cohesiveness and the sense of belonging to a team, all of which may help the organization

to retain better employees. While in theory it seems reasonable to assume that

experiencing help enhances employee retention, there is little evidence of this in the

empirical literature. Chen et al. (1998) reported findings showing first that the level of

altruism was higher in the no turnover than in the turnover condition and second that

altruism was negatively related to turnover intention (correlation matrix, r ¼ 20.15,

p , 0.05). Using regression, Coyne and Ong (2007) found no significant relationship,

while Paille and Grima (2011) found that helping explains the intention to leave the

current job better than the intention to leave the employer.

OCB towards the organization typically reflects civic virtue and sportsmanship. Civic

virtue refers to a sustained interest in the organization, expressed in a variety of ways,

including assiduous and voluntary involvement in representation activities (e.g.

conferences, trade fairs and workshops) and in the defence of the interests, property or

image of the organization. As such, civic virtue depends on active and voluntary

participation and requires individuals who want to be involved, for example, in decision-

making by formulating new ideas, suggesting improvements in seminars or meetings, or

protecting the organization (Organ et al. 2006). In so doing, employees demonstrate a deep

interest in the political life of the organization. To date, with the exception of Coyne and

Ong (2007) and Paille and Grima (2011), very little research has been conducted on the

effect of civic virtue on turnover intentions. While Coyne and Ong (2007) found no

P. Paille4

Dow

nloa

ded

by [

Uni

vers

ite L

aval

] at

08:

55 1

8 Ju

ly 2

012

significant relationship between civic virtue and turnover intentions, they recognized that

features of the sample (size and participants) may minimize the statistical power of their

data. Paille and Grima (2011) found a negative relationship suggesting that civic virtue

affects intention to leave the organization.

Sportsmanship is typically defined as a form of citizenship directed towards the

organization. As reported by Podsakoff and MacKenzie (1997), several studies have found

that sportsmanship is positively related to a number of organizational performance criteria.

Sportsmanship occurs when an employee is willing to avoid voicing complaints about

trivial matters and to set about being an example for others (Organ et al. 2006). In addition,

some studies have argued that sportsmanship reflects the willingness to ‘[stay] with the

organization despite hardships or difficult conditions’ (Coleman and Borman 2000, p. 34).

In so doing, an employee shows a sense of forbearance, enduring individual costs or

frustrations related to his or her work (Organ 1990). Sportsmanship can be used by

employers as an indicator of the degree of individual tolerance requirements to which an

employee is subjected in their work. Chen et al. (1998) reported two interesting findings.

First, the level of sportsmanship was found to be higher among non-quitters than among

quitters, and second sportsmanship was negatively associated with turnover intention

(correlation matrix, r ¼ 20.19, p , 0.01). However, in their subsequent calculations

(i.e. regression), Chen et al. (1998) opted to use a global measure of OCB rather than distinct

forms of OCB. To date, only Coyne and Ong (2007) and Paille and Grima (2011) have used

regression to examine the extent to which sportsmanship contributes to the variance of

intention to leave the organization. Both studies reported similar findings and demonstrated

that sportsmanship was the best predictor. These results are consistent with Chen et al.

(1998) and suggest that sportsmanship reflects a desire to remain in the organization.

As noted above, a significant amount of empirical research has been conducted on the

relationship between OCB and employee retention (actual turnover and turnover

intention). Studies in the area have shown the utility of OCB as a predictor of the decision

to quit. However, intention to leave the organization is only one dimension among others

of turnover cognitions. While researchers have sought to determine how OCB and

intention to leave the organization are related, other aspects of turnover cognitions, such as

intention to search and perceived job alternatives, have been largely overlooked. However,

two studies (Hui et al. 1999; Thau et al. 2004) have provided useful results on the

relationship between perceived job alternatives and OCB. Hui et al. (1999) were the first to

provide a direct assessment of the relationship between perceived job alternatives and

OCB. They suggested that the willingness to perform OCB is a function of perceived job

alternatives (referred to as ‘job mobility’). Employees continue to exhibit OCB when they

perceive that job mobility is reduced. Hui et al. (1999) reported a significant negative

relationship in a Chinese context, suggesting that ‘employees who perceived a vast variety

of outside job opportunities have a lower tendency to exhibit extra-role behaviors’ (p. 14).

More recently, Thau et al. (2004) noted that Hui et al. (1999) made no distinction

between ‘situations in which employees are restricted in their context (e.g. when perceived

ease of finding a new job is low) from situations in which employees face very few

restrictions (e.g. when perceived ease of alternative employment is high)’ (p. 609). To go

one step further, Thau et al. (2004) used attractiveness and social networking to explain the

decision of employees to engage (or not) in OCB. Attractiveness refers to the degree of

dependence of an employee on his or her organization associated with the existence (or not)

of alternative job opportunities outside the organization. Social networking refers to

aspects of the employee’s personal life or professional factors that reduce the desire of the

employee to move to another organization. The decision to engage in OCB is a function of

The International Journal of Human Resource Management 5

Dow

nloa

ded

by [

Uni

vers

ite L

aval

] at

08:

55 1

8 Ju

ly 2

012

both attractiveness and social networking. In a German context, Thau et al. (2004) found

that when employees perceive that alternative employment opportunities are easy to find

(high attractiveness and low social networking), they are less likely to engage in OCB than

employees who perceive that opportunities cannot be easily found (low attractiveness and

high social networking).

The findings by Thau et al. (2004) are consistent with research on job embeddedness

(Lee et al. 2004). Following Yao, Lee, Mitchell, Burton and Sablynski (2004), job

embeddedness refers to ‘the forces that keep a person from leaving his or her job’ (p. 157).

Job embeddedness is a composite construct of on- and off-the-job embeddedness (Mitchell,

Holtom, Lee, Sablynski and Erez 2001). While on-the-job embeddedness refers to a

situation in which an employee remains in his or her organization because of forces related to

the job (i.e. company culture, professional growth and development, etc.), off-the-job

embeddedness refers to a situation in which an employee remains in the organization

because of forces related to the community (i.e. family roots, safe neighbourhood, etc.). Lee

et al. (2004) found that on-the-job embeddedness was significantly predictive of OCB,

whereas off-the-job embeddedness was not.

Research purposes

Although the findings outlined by Hui et al. (1999) and Thau et al. (2004) indicate a

significant shift by showing how perceived job alternatives affect OCB, progress can be

made by taking account of intention to search and intention to leave the organization.

However, as Thau et al. (2004) suggest, perceived stimulating job conditions outside the

organization may be inhibited by personal limitations. In others words, perceived

stimulating and interesting opportunities in the labour market do not mean that individuals

will actively seek for a job. Typically, perceived job alternatives, intention to search and

intention to leave are encompassed in turnover cognitions. According to Sager, Griffeth

and Hom (1998), turnover cognitions reflect ‘a mental decision intervening between an

individual’s attitudes regarding a job and the stay or leave decision’ (p. 255). Previous

research on turnover cognitions indicates that perceived job alternatives affect intention to

leave through intention to search (Mobley 1977). In a replication of the study by Mobley,

Horner and Hollingsworth (1978), Miller, Katerberg and Hulin (1979) found similar

patterns and established the generalizability of the model. However, further research on

different samples was recommended. Although Mobley’s model is an accepted reference

in the turnover literature, research has shown that both the type of employment and the

type of industry may affect relationships between perceived job alternatives, job search

and intention to leave (Mowday, Koberg and MacArthur 1984; Sager et al. 1998).

Jaros, Jermier, Koehler and Sincich (1993) argued that research ‘might theorize that

individuals typically exhibit vague, general orientations and tendencies toward a particular

behavior rather than distinct, sequentially ordered, focused cognitions’ (p. 986). This

implies that employees may decide to withdraw from their current organization without

having developed specific plans. In other words, while it has been shown that perceived

job alternatives increase the probability that an employee will search for a better job and in

turn leave their employer, this sequence needs to be viewed with caution in some settings.

For example, shortage within a particular industry may breach the sequence. Consistent

with Sager et al. (1998), job search may be superfluous in some cases (i.e. nurses, truck

drivers, IT workers, salespeople, social workers, etc.). Employees in these employment

settings may assume that it is easy to find a new job. Since previous research has been

conducted to explain how perceived job alternatives affect OCB and how it is linked to

P. Paille6

Dow

nloa

ded

by [

Uni

vers

ite L

aval

] at

08:

55 1

8 Ju

ly 2

012

intention to leave the organization through intention to search, this study focuses primarily

on effort, following Hui et al. (1999) and Thau et al. (2004), by examining whether forms

of OCB increase or decrease intention to search and in turn affect intention to leave the

organization. The above discussion can be summarized by the following hypotheses:

Hypothesis 1: Perceived job alternatives are negatively related to (a) helping,

(b) altruism, (c) sportsmanship, and (d) civic virtue.

Hypothesis 2: (a) Helping, (b) altruism, (c) sportsmanship, and (d) civic virtue are

negatively related to intention to search.

Hypothesis 3: (a) Helping, (b) altruism, (c) sportsmanship, and (d) civic virtue are

negatively related to intention to leave the organization.

Hypothesis 4: Perceived job alternatives are positively related to intention to search.

Hypothesis 5: Intention to search is positively related to intention to leave the

organization.

The research model is consistent with previous findings showing first that perceived

job alternatives affect forms of OCB (Hui et al. 1999; Thau et al. 2004) and second that

forms of OCB affect intention to leave the organization (Coyne and Ong 2007; Paille and

Grima 2011). Intention to search was added to take account of research on turnover

cognitions showing that perceived job alternatives affect intention to leave through

intention to search (Miller et al. 1979; Bozeman and Perrewe 2001). In short, the research

model examined (1) the paths connecting perceived job alternatives and forms of OCB;

(2) the path connecting perceived job alternatives and intention to search; (3) the paths

connecting forms of OCB and intention to search; (4) the path connecting intention to

search and intention to leave the organization; and (5) the paths connecting forms of OCB

and intention to leave the organization.

Alternative models

Coyne and Ong (2007) suggested that ‘future research would need to employ more

longitudinal designs to examine in more detail whether OCB does predict turnover intention

or whether turnover intention actually predicts OCB behaviour’ (p. 1094). In both studies,

the design was cross-sectional (see below). Therefore, to provide findings indicating that

OCB predicts turnover cognitions rather than findings showing correlations, it was

important to assess the direction of causation. Consistent with Coyne and Ong (2007),

the research model (Model 1) was compared to four alternative models (see Figure 1,

Models 2–5).

Model 2

Model 2 is a variant of the research model (Model 1) and takes account of the results of the

study by Jaros et al. (1993) by dropping the path from intention to search and intention to

leave the organization. Therefore, Model 2 examined (1) the paths connecting perceived

job alternatives and forms of OCB, (2) the paths connecting forms of OCB and intention to

search, and (3) the paths connecting forms of OCB and intention to leave the organization.

Model 3

Model 3 considers forms of OCB as determinants of turnover cognitions. The main

difference with the research model is the causal relationship between perceived job

The International Journal of Human Resource Management 7

Dow

nloa

ded

by [

Uni

vers

ite L

aval

] at

08:

55 1

8 Ju

ly 2

012

alternatives and intention to leave the organization. Previous research found a negative

relationship between forms of OCB and intention to leave the organization (Paille and

Grima 2011) and reported that perceived job alternatives affect the change of employer via

job search (Bozeman and Perrewe 2001). On the basis of these data, it seems plausible to

conceptualize forms of OCB as determinants of turnover cognitions. Therefore, Model 3

Perceived jobalternatives

OCBO(sportsmanshipand civic virtue)

Intent to search Intent to leave

OCBI (altruismand helping)

Model 1(Research model)

Perceived jobalternatives

OCBO(sportsmanshipand civic virtue)

Intent to search Intent to leave

OCBI (altruismand helping)

Model 2 - Dropping thepath from intent tosearch to intent toleave the organization.

Perceivedjob alternatives

OCBO(sportsmanshipand civic virtue)

Intent to search Intent to leave

OCBI(altruism and

helping)

Model 3 - OCBs asdeterminants to turnovercognitions.

Perceived jobalternatives

OCBO(sportsmanshipand civic virtue)

Intent to search Intent to leave

OCBI(altruism and

helping)

Model 5 - OCBs asoutcomes to turnovercognitions.

Perceived jobalternatives

OCBO(sportsmanshipand civic virtue)

Intent to search Intent to leave

OCBI(altruism and

helping)

Model 4 - OCBs asintermediate variablesbetween perceived jobalternatives and intentto search.

Figure 1. Competing models representing the relationships between turnover cognitions andOCBs.

P. Paille8

Dow

nloa

ded

by [

Uni

vers

ite L

aval

] at

08:

55 1

8 Ju

ly 2

012

examined (1) the paths connecting forms of OCB and perceived job alternatives, (2) the

path connecting perceived job alternatives and intention to search, and (3) the path

connecting intention to search and intention to leave the organization.

Model 4

On the basis of previous research (Hui et al. 1999; Thau et al. 2004), the research model

(Model 1) proposes to consider forms of OCB as intermediate variables between perceived

job alternatives and intention to search. On the basis of assumption that perceived job

alternatives affect intention to leave the organization via intention to search (Bozeman and

Perrewe 2001), as a possible competing model, Model 4 examines forms of OCB as

intermediate variables between intention to search and intention to leave the organization.

The rationale is suggested by the following assumptions. Boswell, Zimmerman and

Swider (2012) argued that job search ‘can affect outcomes for the employer’ (p. 147).

Given that as important outcomes OCBs enhance organizational performance (Organ et al.

2006), that low OCBs reflect a propensity to withdraw from the organization (Chen 2005)

and that OCBs are related to intention to leave (Paille and Grima 2011), it seems

reasonable to examine (1) the path connecting perceived job alternatives and intention to

search, (2) the paths connecting intention to search and forms of OCB, and (3) the paths

connecting forms of OCB and intention to leave the organization.

Finally, Model 5 considers forms of OCB as outcomes of turnover cognitions. Griffeth

et al. (2000) showed that the withdrawal process (i.e. perceived job alternatives !

intention to search ! intention to leave) is the best predictor of turnover (not examined in

this paper). Chen et al. (1998) showed that turnover is better predicted by OCB than

turnover intention. These findings suggest that OCB can be a realistic output of the

withdrawal process. Thus, Model 5 examined (1) the path connecting perceived job

alternatives and intention to search, (2) the path connecting intention to search and

intention to leave the organization, and (3) the paths connecting intention to leave the

organization and forms of OCB.

Method

Though convenience samples were used in both Study 1 and Study 2, the samples were

different. Unlike Study 1, Study 2 was based on a sample in a specific industry affected by

shortage (social work).

Participants

Sample 1

For Study 1, a questionnaire was sent via e-mail to 2441 alumni of a Canadian university.

Of the 2441 individuals contacted as part of the survey, 731 responded, yielding a 29.9%

response rate. After reading the request for participation outlining the aims and objectives

of the research and reading the consent form outlining the ethical guidelines, 16

respondents withdrew from the study and opted not to take part in the online survey. Of the

715 completed questionnaires, 64 questionnaires were discarded because of too much

missing data. In total, 651 questionnaires were completed. 69.7% were women and 41.5%

were aged under 40. Of the respondents, 68% had over 15 years of professional experience

and 68% worked in an organization with more than 500 employees. Of the participants,

The International Journal of Human Resource Management 9

Dow

nloa

ded

by [

Uni

vers

ite L

aval

] at

08:

55 1

8 Ju

ly 2

012

92.5% stated that they had good career development prospects in their current

organization.

Sample 2

Study 2 was carried out in collaboration with the professional association of Quebecer

social workers. The decision to collect data using a sample of social workers was based on

the assumption that the retention of social workers is an issue of particular concern not only

for employers but also in view of the significant number of workers leaving the profession

(Wermeling 2009). In total, 230 individuals employed in social work responded to the

survey. After reading the request for participation outlining the aims and objectives of

the research and reading the consent form outlining the ethical guidelines, four respondents

withdrew from the study and opted not to take part in the online survey. The sample

included 226 participants. Of the participants, 88% were women and 53% were aged under

40. Of the respondents, 37.6% had over 15 years of professional experience and 65.6%

worked in an organization with more than 500 employees. Of the participants, 93% stated

that they had good career development prospects in their organization.

Measurement

The same measurements were used in both studies.

OCBs

Using self-report measures, three subscales developed by Podsakoff and MacKenzie

(1994), measuring helping behaviours (six items), civic virtue (three items) and

sportsmanship (four items), were used. The source of OCB measurement is a controversial

topic in the literature. The issue of common variance is the main argument given by the

advocates of other types of report (supervisor or peer). However, many researchers

(Conway and Lance 2010) have tended to minimize the effects of self-report results related

to the use of scales of OCB. Others (e.g. Turnipseed 2002) see the use of methods based on

self-reporting as justified when the investigation is focused on an examination of the links

between psychological variables. As Meierhans, Rietmann and Jonas (2008) argue,

‘the assumption underlying the use of self-reports is that OCB is generally not carried out

in front of the supervisor and is therefore better known to the actor himself’ (p. 139).

The purpose is to examine how OCB and turnover cognitions (psychological variables) are

related. Although the risks that may cause common variance are known, using self-report

is justified in the present study. Because previous empirical research conducted in the

francophone context (Paille 2009) found evidence of a four-factor model (altruism,

helping, sportsmanship and civic virtue), and because this study was also conducted in the

francophone context, similar patterns were logically predicted.

For Study 1, a confirmatory factor analysis found that a four-factor model provides a

better fit, x 2(61, n ¼ 651) ¼ 222.45, p , 0.000, comparative fit index (CFI) ¼ 0.93, non-

normed fit index (NNFI) ¼ 0.92 and root mean square error of approximation

(RMSEA) ¼ 0.06, than a three-factor model, x 2diff(1) ¼ 480.17, p , 0.001, and a

one-factor model, x 2diff(4) ¼ 932.21, p , 0.001. It was concluded that the participants

were able to distinguish the four OCB dimensions of altruism (a ¼ 0.87), helping

(a ¼ 0.79), civic virtue (a ¼ 0.70) and sportsmanship (a ¼ 0.72). Table 1 shows the

items of the four scales.

P. Paille10

Dow

nloa

ded

by [

Uni

vers

ite L

aval

] at

08:

55 1

8 Ju

ly 2

012

Tab

le1

.M

easu

rem

ent

mo

del

for

Stu

dy

1an

dS

tud

y2

.

Stu

dy

1(n

¼6

51

)S

tud

y2

(n¼

22

6)

Lo

ad

ing

sr

CR

AV

EL

oa

din

gs

rC

RA

VE

Alt

ruis

m0

.98

0.9

30

.97

0.9

40

.78

0.8

8I

giv

em

yti

me

toh

elp

coll

eag

ues

wh

oh

ave

wo

rk-r

elat

edp

rob

lem

s0

.86

30

.89

7

Iw

illi

ng

tota

ke

tim

eo

ut

of

my

ow

nb

usy

sch

edu

leto

hel

pn

ewco

llea

gu

es0

.90

10

.64

6

Hel

pin

g0

.97

0.9

10

.92

0.9

20

.87

0.8

1I

tak

est

eps

totr

yto

pre

ven

tp

rob

lem

sw

ith

oth

erp

erso

nn

elin

the

com

pan

y0

.70

30

.91

0

Iac

tas

a‘p

eace

mak

er’

wh

eno

ther

sin

the

com

pan

yh

ave

dis

agre

emen

ts0

.85

70

.80

9

Iam

ast

abil

izin

gin

flu

ence

inth

eco

mp

any

wh

end

isse

nti

on

occ

urs

0.8

61

0.5

65

I‘t

ou

chb

ase’

wit

ho

ther

sb

efo

rein

itia

tin

gac

tio

ns

that

mig

ht

affe

ctth

em(r

emo

ved

inS

tud

y1

and

Stu

dy

2)

––

Civ

icvi

rtu

e0

.93

0.7

90

.83

0.8

90

.86

0.8

0I

atte

nd

fun

ctio

ns

that

are

no

tre

qu

ired

bu

th

elp

the

com

pan

yim

age

0.7

16

0.8

11

Iat

ten

din

form

atio

nse

ssio

ns

that

emp

loy

eear

een

cou

rag

edb

ut

no

tre

qu

ired

toat

ten

d0

.71

90

.77

8

Iac

tiv

ely

par

tici

pat

esin

com

pan

ym

eeti

ng

s0

.56

10

.50

1S

po

rtsm

an

ship

(all

sco

res

reve

rsed

)0

.88

0.7

10

.72

0.7

80

.78

0.7

0I

con

sum

ea

lot

of

tim

eco

mp

lain

ing

abo

ut

triv

ial

mat

ters

0.5

63

0.8

05

Ite

nd

tom

ake

‘mo

un

tain

so

ut

the

mo

leh

ills

’0

.51

10

.67

4I

alw

ays

focu

so

nw

hat

isw

ron

gw

ith

my

situ

atio

nra

ther

than

the

po

siti

ve

sid

eo

fit

0.6

80

0.5

89

Ial

way

sfi

nd

fau

ltw

ith

wh

atth

eco

mp

any

isd

oin

g(r

emo

ved

inS

tud

y1

and

Stu

dy

2)

––

(Co

nti

nu

ed)

The International Journal of Human Resource Management 11

Dow

nloa

ded

by [

Uni

vers

ite L

aval

] at

08:

55 1

8 Ju

ly 2

012

Tab

le1

–co

nti

nu

ed

Stu

dy

1(n

¼6

51

)S

tud

y2

(n¼

22

6)

Lo

ad

ing

sr

CR

AV

EL

oa

din

gs

rC

RA

VE

Inte

nt

tole

ave

the

org

an

iza

tio

n0

.97

0.9

60

.93

0.9

60

.91

0.8

9T

her

eis

ag

oo

dch

ance

that

Iw

ill

leav

eth

iso

rgan

izat

ion

inth

en

ext

yea

r0

.91

60

.91

1

Ifr

equ

entl

yth

ink

of

leav

ing

this

org

aniz

atio

n0

.85

10

.84

3I

wil

lp

rob

ably

loo

kfo

ra

new

org

aniz

atio

nin

the

nex

ty

ear

0.9

67

0.8

88

Inte

nt

tose

arc

h0

.95

0.9

30

.88

0.9

30

.94

0.8

3I

oft

ense

ekin

form

atio

nab

ou

to

ther

job

po

ssib

ilit

ies

0.9

08

0.9

81

Io

ften

loo

kat

new

spap

erad

sfo

rn

ewjo

bs

0.8

33

0.8

21

Iso

met

imes

foll

ow

up

on

job

lead

sI’

ve

hea

rdab

ou

t0

.82

80

.82

9P

erce

ived

job

alt

ern

ati

ves

0.9

50

.89

0.8

80

.93

0.8

90

.83

Itis

po

ssib

lefo

rm

eto

fin

da

bet

ter

job

than

the

on

eI

hav

en

ow

0.6

85

0.6

25

Acc

epta

ble

job

sca

nal

way

sb

efo

un

d0

.87

00

.91

6N

od

ou

bt

inm

ym

ind

that

Ica

nfi

nd

ajo

bth

atis

atle

ast

asg

oo

das

the

on

eI

no

wh

ave

0.7

93

0.8

16

No

te:r,

Jore

sko

g’s

r;

CR

,co

mposi

tere

liab

ilit

y;

AV

E,

aver

age

var

iance

extr

acte

d.

P. Paille12

Dow

nloa

ded

by [

Uni

vers

ite L

aval

] at

08:

55 1

8 Ju

ly 2

012

For Study 2, the confirmatory factor analysis indicated that the four-factor model

provides a better fit, x 2(61, n ¼ 226) ¼ 120.63, p , 0.000, CFI ¼ 0.92, GFI ¼ 0.92 and

RMSEA ¼ 0.06, than the three-factor model, x 2diff(1) ¼ 76.51, p , 0.001, and the

one-factor model, x 2diff(4) ¼ 303.87, p , 0.001. It was concluded that social workers were

able to distinguish the four OCB dimensions of altruism (a ¼ 0.72), helping (a ¼ 0.81),

civic virtue (a ¼ 0.70) and sportsmanship (a ¼ 0.69).

Turnover cognitions

The three-item scale developed by Lichtenstein, Alexander, McCarthy and Wells (2004) was

used to measure intention to quit (Study 1, a ¼ 0.93; Study 2, a ¼ 0.91). The three-item

scale drawn from Peters, Jackofsky and Salter (1981) was used to measure perceived job

alternatives (Study 1, a ¼ 0.82; Study 2, a ¼ 0.82). The three-item scale developed by

Peters et al. (1981) was used to measure intention to search (Study 1, a ¼ 0.89; Study 2,

a ¼ 0.90). Table 1 shows the items of the three scales. Bozeman and Perrewe (2001)

indicate that ‘a number of studies have shown that turnover intention measures are not well

differentiated from measures of other closely associated precursors to actual turnover,

specifically, from thoughts of quitting and intentions to search for another job with a different

organization’ (p. 162). To avoid spurious findings, a confirmatory factor analysis was

conducted to determine whether the items of each measure load on their appropriate factor.

For Study 1, the confirmatory factor analysis results indicate that the three-factor model

fits the data better, x 2(24, n ¼ 651) ¼ 97.65, p , 0.000, CFI ¼ 0.98, NNFI ¼ 0.97 and

RMSEA ¼ 0.06, than (1) a model that combines two factors into one factor, intention to

search and intention to leave, x 2diff(2) ¼ 663.37, p , 0.001, (2) a model that combines two

factors, intention to search and perceived job alternatives, x 2diff(2) ¼ 712.08, p , 0.001, (3)

a model that combines two factors, intention to quit and perceived job alternatives,

x 2diff(2) ¼ 656.95, p , 0.001, and (4) a model that combines in one factor intention to

search, perceived job alternatives and intention to leave, x 2diff(4) ¼ 1802.53, p , 0.001.

For Study 2, the confirmatory factor analysis results indicate that the three factors

provided an excellent fit, x 2(22, n ¼ 226) ¼ 47.29, p , 0.001, CFI ¼ 0.98, NNFI ¼ 0.97

and RMSEA ¼ 0.07. The three-factor model fitted the data better than (1) a model that

combined two factors into one factor, intention to search and intention to leave

x 2diff(4) ¼ 342.46, p , 0.001, (2) a model that combined two factors into one factor,

intention to search and perceived job alternatives, x 2diff(4) ¼ 456.81, p , 0.001, (3) a

model that combined two factors into one factor, intention to quit and perceived job

alternatives, x 2diff(2) ¼ 378.97, p , 0.001, and (4) a model that combined in one factor

intention to search, perceived job alternatives and intention to quit x 2diff(7) ¼ 675.21,

p , 0.001.

It was concluded that the discriminant validity of perceived job alternatives, intention

to search and intention to leave was supported in both studies.

For both studies, all items were measured using a seven-point Likert scale with

responses ranging from 1 (disagree) to 7 (agree).

Results

For both studies, the recommendations outlined by Anderson and Gerbing (1988) were

followed. A two-stage process was followed to analyse the data. First, the measurement

model was examined to evaluate the independence of the constructs. Second, structural

equation modelling (SEM) was used to analyse the data.

The International Journal of Human Resource Management 13

Dow

nloa

ded

by [

Uni

vers

ite L

aval

] at

08:

55 1

8 Ju

ly 2

012

Measurement model (Study 1 and Study 2)

A measurement model consisting of seven factors (helping, altruism, civic virtue,

sportsmanship, intention to search, perceived job alternatives and intention to quit) fitted

the data well [Study 1, x 2(193, n ¼ 651) ¼ 494.67, p , 0.000, CFI ¼ 0.96, NNFI ¼ 0.95

and RMSEA ¼ 0.05; Study 2, x 2(190, n ¼ 226) ¼ 284.42, p , 0.000, CFI ¼ 0.96,

NNFI ¼ 0.95 and RMSEA ¼ 0.04].

For both studies, Table 1 shows the composite reliability (CR), estimating the extent to

which a set of latent construct indicators shares in their measurement of a construct;

average variance extracted (AVE), representing the ratio of the total variance due to the

latent variable; and Joreskog’s r for all constructs of the sample. Hair, Anderson, Tatham

and Black (1998) recommend a CR above the 0.70 threshold and an AVE above the 0.50

threshold. Since the values respected the recommended cut-off in both studies, the

measurement model may be said to provide evidence of the reliability, convergent validity

and discriminant validity of the measures.

SEM results (Study 1 and Study 2)

Following Anderson and Gerbing (1988), the measurement model was used at this stage.

For both studies, Model 1 (the research model) fitted the data well [Study 1, x 2(198,

n ¼ 651) ¼ 633.16, p , 0.000, CFI ¼ 0.93, NNFI ¼ 0.92, and RMSEA ¼ 0.06; Study 2,

x 2(195, n ¼ 226) ¼ 373.69, p , 0.000, CFI ¼ 0.93, NNFI ¼ 0.91 and RMSEA ¼ 0.06].

Table 2 summarizes the results of the comparison of competing models. The x 2

difference test suggested that the research model (Model 1) improves the fit compared to

Model 2 in both studies [Study 1, x 2diff(5) ¼ 271.84, p , 0.001; Study 2, x 2

diff(0) ¼ 20.40,

p , 0.001], Model 3 [Study 1, x 2diff(4) ¼ 70.97, p , 0.001; Study 2, x 2

diff(4) ¼ 28.61,

p , 0.001], Model 4 [Study 1, x 2diff(3) ¼ 390.02, p , 0.001; Study 2, x 2

diff(1) ¼ 50.43,

p , 0.001], and Model 5 [Study 1, x 2diff(4) ¼ 127.54, p , 0.001; Study 2,

x 2diff(4) ¼ 41.61, p , 0.001]. Therefore, the more parsimonious research model

(Model 1) was used to test the hypotheses.

Research hypotheses (Study 1 and Study 2)

Table 3 provides an overview of the results of both studies.

Hypotheses 1(a)–(d) predicted negative relationships between perceived job

alternatives and forms of OCB. The factor perceived job alternatives was negatively

related to helping (Study 1, b ¼ 20.281, t-value ¼ 25.944, p ¼ 0.000; Study 2,

b ¼ 20.162, t-value ¼ 22.155, p ¼ 0.031), altruism (Study 1, b ¼ 20.294,

t-value ¼ 26.183, p ¼ 0.000; Study 2, b ¼ 20.220, t-value ¼ 22.992, p ¼ 0.003),

sportsmanship (Study 1, b ¼ 20.490, t-value ¼ 27.419, p ¼ 0.000; Study 2,

b ¼ 20.493, t-value ¼ 24.540, p ¼ 0.000) and civic virtue (Study 1, b ¼ 20.350,

t-value ¼ 26.482, p ¼ 0.000; Study 2, b ¼ 20.347, t-value ¼ 24.018, p ¼ 0.000).

Thus, for both studies, the results support Hypotheses 1(a)–(d).

Hypotheses 2(a)–(d) predicted negative relationships between forms of OCB and

intention to search. In both studies, helping was not related to intention to search (Study 1,

b ¼ 0.043, n.s.; Study 2, b ¼ 0.048, n.s.). Altruism was related to intention to search in

Study 1 but not in the predicted direction (b ¼ 0.112, t-value ¼ 2.403, p ¼ 0.026), but

was not related to intention to search in Study 2 (b ¼ 20.079, n.s.). In both studies,

sportsmanship (Study 1, b ¼ 20.210, t-value ¼ 23.607, p ¼ 0.000; Study 2,

b ¼ 20.383, t-value ¼ 23.779, p ¼ 0.000) was negatively related to intention to search.

P. Paille14

Dow

nloa

ded

by [

Uni

vers

ite L

aval

] at

08:

55 1

8 Ju

ly 2

012

Tab

le2

.C

om

par

iso

no

fco

mp

etin

gm

od

els.

Stu

dy

1(n

¼6

51

)S

tud

y1

(n¼

22

6)

Co

mp

etin

gm

od

els

x2

df

RM

SE

AC

FI

NN

FI

x2

df

RM

SE

AC

FI

NN

FI

Mo

del

1–

Res

earc

hm

od

el6

33

.16

**

*1

98

0.0

60

.93

0.9

23

73

.69

**

*1

95

0.0

60

.93

0.9

1M

od

el2

–D

rop

pin

gth

ep

ath

fro

min

ten

tto

sear

chto

inte

nt

tole

ave

the

org

aniz

atio

n9

05

.00

**

20

30

.08

0.9

10

.90

39

4.0

9*

*1

95

0.0

80

.93

0.9

0

Mo

del

3–

OC

Bs

asd

eter

min

ants

totu

rno

ver

cog

nit

ion

s7

04

.13

**

20

20

.08

0.9

30

.92

40

2.3

0*

*1

99

0.0

60

.93

0.9

0M

od

el4

–O

CB

sas

inte

rmed

iate

var

iab

les

bet

wee

np

erce

ived

job

alte

rnat

ives

and

inte

nt

tose

arch

10

23

.18

**

20

10

.11

0.8

80

.87

42

4.1

2*

*1

96

0.1

10

.89

0.8

7

Mo

del

5–

OC

Bs

aso

utc

om

esto

turn

ov

erco

gn

itio

ns

76

0.7

0*

*2

02

0.1

10

.85

0.8

54

15

.30

**

19

90

.11

0.8

80

.85

No

te:

**

*p,

0.0

00

;*

*p,

0.0

01

.

The International Journal of Human Resource Management 15

Dow

nloa

ded

by [

Uni

vers

ite L

aval

] at

08:

55 1

8 Ju

ly 2

012

Tab

le3

.P

ath

anal

ysi

sre

sult

s.

Stu

dy

1(n

¼6

51

)S

tud

y2

(n¼

22

6)

Res

earc

hh

ypo

thes

esS

tan

da

rdiz

ed

pa

thco

effici

ent

( b)

t-va

lue

pS

tan

da

rdiz

ed

pa

thco

effici

ent

(b)

t-va

lue

p

H1

(a)

PJA

!H

elpin

g2

0.2

81

25

.944

0.0

00

Su

ppo

rted

20

.162

22

.15

50

.003

Su

pp

ort

edH

1(b

)P

JA!

Alt

ruis

m2

0.2

94

26

.183

0.0

00

Su

ppo

rted

20

.220

22

.99

20

.031

Su

pp

ort

edH

1(c

)P

JA!

Sp

ort

sman

ship

20

.49

02

7.4

19

0.0

00

Su

ppo

rted

20

.493

24

.54

00

.000

Su

pp

ort

edH

1(d

)P

JA!

Civ

icv

irtu

e2

0.3

50

26

.482

0.0

00

Su

ppo

rted

20

.347

24

.01

80

.000

Su

pp

ort

edH

2(a

)H

elpin

g!

Inte

nt

tose

arch

0.0

43

0.8

23

n.s

.N

on

-su

pp

ort

ed0

.048

0.8

88

n.s

.N

on

-su

pp

ort

edH

2(b

)A

ltru

ism

!In

ten

tto

sear

ch0

.11

22

.403

0.0

26

No

n-s

up

po

rted

20

.079

20

.93

2n

.s.

No

n-s

upp

ort

edH

2(c

)S

po

rtsm

ansh

ip!

Inte

nt

tose

arch

20

.21

02

3.6

07

0.0

00

Su

ppo

rted

20

.383

23

.77

90

.000

Su

pp

ort

edH

2(d

)C

ivic

vir

tue

!In

ten

tto

sear

ch2

0.2

33

23

.763

0.0

00

Su

ppo

rted

20

.010

20

.23

2n

.s.

No

n-s

upp

ort

edH

3(a

)H

elpin

g!

Inte

nt

tole

ave

0.0

75

2.0

89

0.0

37

No

n-s

up

po

rted

0.0

50

0.8

89

n.s

.N

on

-su

pp

ort

edH

3(b

)A

ltru

ism

!In

ten

tto

leav

e2

0.0

79

22

.044

0.0

47

Su

ppo

rted

20

.060

20

.90

4n

.s.

No

n-s

upp

ort

edH

3(c

)S

po

rtsm

ansh

ip!

Inte

nt

tole

ave

20

.19

42

4.6

36

0.0

00

Su

ppo

rted

20

.255

23

.02

10

.003

Su

pp

ort

edH

3(d

)C

ivic

vir

tue

!In

ten

tto

leav

e0

.01

80

.243

n.s

.N

on

-su

pp

ort

ed2

0.2

01

22

.93

20

.003

Su

pp

ort

edH

(4)

PJA

!In

ten

tto

sear

ch0

.14

62

.464

0.0

14

Su

ppo

rted

0.1

05

1.1

09

n.s

.N

on

-su

pp

ort

edH

(5)

Inte

nt

tose

arch

!In

ten

tto

leav

e0

.64

41

6.8

42

0.0

00

Su

ppo

rted

0.3

16

4.5

02

0.0

00

Su

pp

ort

ed

Note

:P

JA,

per

ceiv

edjo

bal

tern

ativ

es;

n.s

.,non-s

ignifi

cant.

P. Paille16

Dow

nloa

ded

by [

Uni

vers

ite L

aval

] at

08:

55 1

8 Ju

ly 2

012

Finally, civic virtue was related to intention to search in Study 1 (b ¼ 20.233,

t-value ¼ 23.763, p ¼ 0.000), but not in Study 2 (b ¼ 20.010, n.s.). Therefore, while

Hypotheses 2(a) and (b) were not supported in Studies 1 and 2, Hypothesis 2(c) was

supported in both studies; finally, Hypothesis 2(d) was supported in Study 1 but not in

Study 2.

Hypotheses 3(a)–(d) predicted negative relationships between OCBs and intention to

leave the organization. Helping was found to be related to intention to leave in Study 1 but

not in the predicted direction (b ¼ 0.075, t-value ¼ 2.089, p ¼ 0.037), but was not related

to intention to leave in Study 2 (b ¼ 0.050, n.s.). Altruism was related to intention to leave

in Study 1 (b ¼ 20.079, t-value ¼ 22.044, p ¼ 0.047), but not in Study 2 (b ¼ 20.060,

n.s.). In both studies, sportsmanship was negatively related to intention to leave (Study 1,

b ¼ 20.194, t-value ¼ 24.636, p ¼ 0.000; Study 2, b ¼ 20.255, t-value ¼ 23.021,

p ¼ 0.003). Finally, civic virtue was not related to intention to leave in Study 1 (b ¼ 0.018,

n.s.), but was related to intention to leave in Study 2 (b ¼ 20.201, t-value ¼ 22.932,

p ¼ 0.003). Therefore, while Hypothesis 3(a) was not supported in Studies 1 and 2,

Hypothesis 2(b) was supported in Study 1 but not in Study 2; Hypothesis 3(c) was

supported in both studies; and Hypothesis 3(d) was not supported in Study 1, but was

supported in Study 2.

Hypothesis 4 predicted a positive relationship between perceived job alternatives and

intention to search. While perceived job alternatives was found to be positively related to

intention to search in Study 1 (b ¼ 0.146, t-value ¼ 2.464, p ¼ 0.014), no relationship

was found in Study 2 (b ¼ 0.105, n.s.). Therefore, the results only support Hypothesis 4 in

Study 1.

Hypothesis 5 predicted a positive relationship between intention to search and

intention to leave. In both studies, intention to search and intention to leave were found to

be positively related (Study 1, b ¼ 0.644, t-value ¼ 16.842, p ¼ 0.000; Study 2,

b ¼ 0.316, t-value ¼ 4.502, p ¼ 0.000). Therefore, the results of both studies support

Hypothesis 5.

Discussion

Findings

Consistent with previous research (Hui et al. 1999; Thau et al. 2004), both Study 1 and

Study 2 found negative relationships between perceived job alternatives and OCB.

However, the distinctions drawn between different forms of OCB extended the findings of

previous studies by providing additional data. Hui et al. (1999) and Thau et al. (2004)

computed a mean score for OCB. As a result, it was difficult to determine, for instance,

whether perceived job alternatives had a greater effect on OCB towards individuals than

on OCB towards the organization. The findings of both studies suggest the same pattern.

Perceived job alternatives explained OCB towards the organization (sportsmanship and

civic virtue) better than OCB towards individuals (helping and altruism). Hui et al. (1999)

indicate that ‘employees who perceive low job mobility may continue their OCB so as to

protect their investment in the organization’ (p. 16). Given that the variance explained for

sportsmanship (Study 1: 23.1%; Study 2: 24.4%) was higher than civic virtue (Study 1:

11.6%; Study 2: 11.4%), the results suggest that a favourable job market (high job

alternatives) increases the likelihood, first, that employees will be unwilling to tolerate

excessive demands in their workplace and, second, that employees will be unwilling to

protect the interests of their employer. In other words, the lack of sportsmanship or/and the

lack of interest in the political dimensions of the organization (relating to governance) may

The International Journal of Human Resource Management 17

Dow

nloa

ded

by [

Uni

vers

ite L

aval

] at

08:

55 1

8 Ju

ly 2

012

be perceived as an indication that employees are engaged in a withdrawal process. In

addition, the perception of job alternatives has a lesser impact on OCB towards

individuals. The variance explained for helping (Study 1: 7.9%; Study 2: 2.6%) and

altruism (Study 1: 8.6%; Study 2: 5.4%) was low. The data suggest that job alternatives do

not have a significant effect on investment between individuals.

The positive relationships between helping and intention to leave and between altruism

and intention to search are unexpected. While no relationships were found in Study 2, in

Study 1 helping was found to be positively related to intention to leave the organization,

while altruism was positively related to intention to search. The findings suggest that the

greater the level of help and altruism among employees, the more likely they are to leave

their employer. Because this study focused solely on the relationship between OCB and

turnover cognitions and no other variable was used, these findings are difficult to interpret.

However, two conflicting explanations seem plausible. First, for reasons unknown, it is

possible that reciprocity is not a common standard for participants in Study 1. In this case,

the perception of a low degree of cohesiveness among employees can feed the propensity

to quit. Second, the evidence suggesting that helping and altruism were positively related

to, respectively, intention to quit and intention to search is consistent with Mossholder et al.

(2005), who argued that ‘individuals view their interpersonal citizenship behaviour as an

investment that increases their value to an organization and profession’ (p. 610). Helping

encourages knowledge sharing, sharing the tricks of the trade between employees, and the

improvement of technical skills (Organ et al. 2006). Therefore, it is conceivable that

the improvement of work practices or the acquisition of new skills will encourage an

employee to gauge his or her value on the labour market. The identification by the

employees of their own value on the labour market does not necessarily mean that they

will leave their employer. Employees may use this information to negotiate with their

employer in order to improve their work conditions.

Following Coyne and Ong (2007) and Paille and Grima (2011), the results of this study

(Studies 1 and 2) indicate that sportsmanship was the most important form of OCB for

explaining intention to leave the organization. Therefore, the study extends the findings of

previous studies. While the data in Study 1 indicated that sportsmanship only explains

intention to leave, in Study 2 the results showed that sportsmanship explains intention to

search better than intention to leave. Since a high level of sportsmanship indicates that the

employee is engaged in his or her work and accepts inconveniences linked to demands

such as occasional extra work, the findings of this study suggest that employees do not plan

to leave their organization in order to find a job with less hardship. The data suggest that in

cases of low levels of sportsmanship, and in the event of good employment prospects

outside their current organization (high job alternatives), employees plan to quit before

seeking a new job (Study 1) or seek a new job before planning to quit (Study 2).

Participants in both studies indicated that they had opportunities for further career

development in their current organization. Unfortunately, the items measuring intention to

search are not sufficiently precise to determine whether employees were seeking

employment in their current organization.

In previous research, the relationships between civic virtue and intention to quit were

controversial. Paille and Grima (2011) found negative relationships between civic virtue

and intention to quit, unlike Coyne and Ong (2007). Unfortunately, since the findings of

Study 1 were consistent with that of Coyne and Ong (2007) and the findings of Study 2

were consistent with that of Paille and Grima (2011), the issue surrounding the conflicting

empirical findings of previous studies will not be resolved by this study. However, the use

of perceived job alternatives and intention to search provide additional data that improve

P. Paille18

Dow

nloa

ded

by [

Uni

vers

ite L

aval

] at

08:

55 1

8 Ju

ly 2

012

the interpretation of the findings. In Study 1, although civic virtue did not affect intention

to leave, it provided a better explanation of intention to search than perceived job

alternatives. This may depend on the extent to which an employee is involved in the

governance process. High levels of civic virtue mean that employees are involved or

interested in the political dimension of their organization, such as participating in

organizational governance and protecting the image and reputation of the organization.

Employees exhibiting high levels of civic virtue demonstrate a willingness to remain in the

organization. Furthermore, as suggested by Study 1, an employee displaying low levels of

civic virtue may consider employment alternatives without planning to leave his or her

current organization. Study 2 produced different results. Although civic virtue does not

affect intention to search, it explains perceived job alternatives better than intention to

leave. The findings suggest that high levels of civic virtue create a sense of responsibility

towards the organization that appears to be incompatible with the desire to leave the

employer in the near future.

Finally, perceived job alternatives negatively affect OCB among German (Thau et al.

2004), Chinese (Hui et al. 1999) and French–Canadian employees (the present study).

The results of Studies 1 and 2 highlight different relationships between OCB and intention

to search and OCB and intention to leave the organization. The findings suggest that one

part of the research model appears to be generalizable, while the other part appears to be

explained by the context of employment.

Practical implications

The results of this study have useful implications for organizations. Sportsmanship refers

to the willingness to remain in the organization despite hardships (Coleman and Borman

2000) and provides useful information about the propensity to leave the organization.

The findings of this study suggest that managers receiving growing complaints related to

difficult work conditions (low sportsmanship) may need to conclude that an individual

threshold has been reached. Therefore, to retain the best employees, managers need to take

appropriate measures to reduce difficult work conditions. The literature on support

provides useful findings. Overall, employees may receive support from many people both

outside (e.g. spouse and friends) and inside the organization (e.g. top management,

immediate superior and co-workers). Although employees may receive support from a

range of sources in the workplace, previous studies have highlighted the critical role of the

immediate superior. The feeling of being supported by the superior enhances employee

retention more than the feeling of being supported by the organization (Eisenberger,

Stinglhamber, Vandenberghe, Sucharski and Rhoades 2002) or by co-workers (Ng and

Sorensen 2008). Thus, managers demonstrate their concern for the health of their

employees, which, in turn, improves employee retention.

Limitations of the study and future research

Despite its contribution to the literature, this study has a number of limitations. First, the

results are based on a single sample and a cross-sectional research design. Therefore,

the data should be interpreted with caution. The results were obtained by collecting data

from two independent samples, following the recommendations given by Murphy (1983).

However, in the design chosen for this study, a specific limitation relating to the difficulty

of inferring causality needs to be acknowledged (Bobko and Stone-Romero 1998).

A duplication of the study using a longitudinal design should mitigate this limitation.

The International Journal of Human Resource Management 19

Dow

nloa

ded

by [

Uni

vers

ite L

aval

] at

08:

55 1

8 Ju

ly 2

012

Second, because Chen (2005) reported that OCB explains more variance of actual turnover

than turnover intentions, it might be objected that the findings should be viewed with

caution. However, research using objective measures of turnover provides guidance for

people who were no longer members of the organization at the time of the investigation.

In other words, this amounts to highlighting the measures that should have been taken by

managers to retain employees before employees definitively left the job. As suggested

by Holtom, Mitchell, Lee and Eberly (2008), preference was therefore given to an analysis

of the variables that incite individuals to remain in their current job or that reduce their

intention to leave. Therefore, a measure of intent appears to be more appropriate. Finally,

although the rating source of OCB remains an object of controversy (Vandenberg, Lance

and Taylor 2005) and appears to be a methodological issue that is difficult to resolve, some

readers may feel that the use of self-reporting measures of OCB entails a significant

limitation. Because the data were collected using self-reported measures, common method

variance bias could lead to an overestimation of the results, particularly as a result of the

phenomenon of social desirability bias (Spector 1987). However, as indicated by Schnake

(1991), ‘supervisor ratings may be biased as a result of halo, or deficient because

citizenship is so difficult to observe’ (p. 741). Van Dyne and Cummings (1990) noted that

both self-report and supervisor ratings of OCB have significant weaknesses (cited by

Schnake 1991). Finally, the measurement of OCB (self-reports vs. supervisor- or

colleague-ratings) entails a number of unresolved issues.

Future research will need to take account of the limitations outlined above. In addition,

future research might usefully focus specifically on sportsmanship. First, based on the

results of this study, it was impossible to determine under what circumstances or

conditions sportsmanship might represent the variable providing the best prediction of the

indicators used to study turnover cognitions (Study 2). Future research could test several

variables known for generating greater pressure at work in order to identify the impact on

sportsmanship. Second, the literature has tended to overlook the extent to which

sportsmanship impacts on the decision to leave. As noted above, many industries are

affected by shortages. In industries where organizations are affected by manpower

shortages, an understanding of variables affecting sportsmanship and in turn the decision

to quit would, in all likelihood, help to improve management practices.

To conclude, the primary objective of this study was to improve our knowledge of the

relationships between OCB and turnover cognitions. The findings indicate that

sportsmanship is the form of citizenship with the greatest predictive power. However,

despite the contributions of this study, further research is needed to broaden the

investigation of the relationship between OCB and retention in the workplace.

References

Anderson, J.C., and Gerbing, D.W. (1988), ‘Structural Equation Modeling in Practice: A Review andRecommended Two-Step Approach,’ Psychological Bulletin, 103, 411–423.

Berry, C.M., Lelchook, A.M., and Clark, M.A. (2011), ‘A Meta-Analysis of the InterrelationshipsBetween Employee Lateness, Absenteeism, and Turnover: Implications for Models ofWithdrawal Behavior,’ Journal of Organizational Behavior, doi: 10.1002/job.778.

Bobko, P., and Stone-Romero, E.F. (1998), ‘Meta-Analysis may be Another Useful Research Tool,but It is not a Panacea,’ in Research in Personnel and Human Resources Management, ed.G.R. Ferris, Stamford, CT: JAI Press, pp. 359–397.

Boswell, W.R., Zimmerman, R.D., and Swider, B.W. (2012), ‘Employee Job Search: Toward anUnderstanding of Search Context and Search Objectives,’ Journal of Management, 38, 1,129–163.

P. Paille20

Dow

nloa

ded

by [

Uni

vers

ite L

aval

] at

08:

55 1

8 Ju

ly 2

012

Bozeman, D.P., and Perrewe, P.L. (2001), ‘The Effect of Item Content Overlap on OrganizationalCommitment Questionnaire-Turnover Cognitions Relationships,’ Journal of Applied Psychol-ogy, 86, 1, 161–173.

Chen, X.-P. (2005), ‘Organizational Citizenship Behavior: A Predictor of Employee VoluntaryTurnover,’ in Handbook of Organizational Citizenship Behavior: A Review of ‘Good Solder’Activity in Organizations, ed. D.L. Turnipseed, New York: Nova Science, pp. 435–454.

Chen, X.-P., Hui, C., and Sego, D.J. (1998), ‘The Role of Organizational Citizenship Behavior inTurnover: Conceptualization and Preliminary Tests of Key Hypotheses,’ Journal of AppliedPsychology, 83, 6, 922–931.

Cirka, C.C. (2005), ‘When Actions Speak as Loudly as Words: Autonomy Support, PsychologicalEmpowerment and Organizational Citizenship Behavior,’ in Handbook of OrganizationalCitizenship Behavior: A Review of ‘Good Solder’ Activity in Organizations, ed. D.L. Turnipseed,New York: Nova Science, pp. 291–328.

Coleman, V.I., and Borman, W.C. (2000), ‘Investigating the Underlying Structure of the CitizenshipPerformance Domain,’ Human Resource Management Review, 10, 25–44.

Conway, J.M., and Lance, C.E. (2010), ‘What Reviewers Should Expect from Authors RegardingCommon Method Bias in Organizational Research,’ Journal of Business and Psychology,25, 325–334.

Coyne, I., and Ong, T. (2007), ‘Organizational Citizenship Behaviour and Turnover Intention:A Cross-Cultural Study,’ International Journal of Human Resource Management, 18, 6,1085–1097.

Dalton, D.R., and Mesch, D.J. (1991), ‘On the Extent and Reduction of Avoidable Absenteeism: AnAssessment of Absence Policy Provisions,’ Journal of Applied Psychology, 76, 6, 810–817.

Eisenberger, R., Stinglhamber, F., Vandenberghe, C., Sucharski, I.L., and Rhoades, L. (2002),‘Perceived Supervisor Support: Contributions to Perceived Organizational Support andEmployee Retention,’ Journal of Applied Psychology, 87, 3, 565–573.

Ferrari, J.M. (1992), ‘Procrastination in the Workplace: Attributions for Failure among Individualswith Similar Behavioral Tendencies,’ Personality and Individual Differences, 13, 3, 315–319.

Gouldner, A.W. (1960), ‘The Norm of Reciprocity: A Preliminary Statement,’American Sociological Review, 25, 2, 161–169.

Griffeth, R., Hom, P., and Gaertner, S. (2000), ‘A Meta-Analysis of Antecedents and Correlates ofEmployee Turnover: Update, Moderator Tests, and Research Implications for the NextMillennium,’ Journal of Management, 26, 3, 463–488.

Gupta, N., and Jenkins, G.D. (1983), ‘Tardiness as a Manifestation of Employee Withdrawal,’Journal of Business Research, 11, 1, 61–75.

Hair, J.F., Anderson, R.E., Tatham, R.L., and Black, W.C. (1998), Multivariate Data Analysis(5th ed.), Upper Saddle River, NJ: Prentice-Hall.

Harrison, D.A., Newman, D.A., and Roth, P.L. (2006), ‘How Important are Job Attitudes?Meta-Analytic Comparisons of Integrative Behavioral Outcomes and Time Sequences,’Academy of Management Journal, 49, 2, 305–325.

Holtom, B.C., Mitchell, T.R., Lee, T.W., and Eberly, M.B. (2008), ‘Turnover and RetentionResearch: A Glance at the Past, Review of the Present, and a Venture into the Future,’The Academy of Management Annals, 1, 231–274.

Hom, P., and Griffeth, R. (1995), Employee Turnover, Cincinnati, OH: South-Western.Hui, C., Law, K.S., and Chen, Z.X. (1999), ‘A Structural Equation Model of the Effects of Negative

Affectivity, Leader–Member Exchange, and Perceived Job Mobility on In-Role and Extra-RolePerformance: A Chinese Case,’ Organizational Behavior and Human Decision Processes, 77, 1,3–21.

Jaros, S.J., Jermier, J.M., Koehler, J.W., and Sincich, T. (1993), ‘Effects of Continuance, Affective,and Moral Commitment on the Withdrawal Process: An Evaluation of Eight Structural EquationModels,’ The Academy of Management Journal, 36, 5, 951–995.

Johns, G. (2002), ‘The Psychology of Lateness, Absenteeism, and Turnover,’ in Handbook ofIndustrial, Work and Organizational Psychology (Vol. 2), eds. N. Anderson, D. Ones,H. Sinangil and C. Viswesvaran, Thousand Oaks, CA: Sage, pp. 232–252.

Koys, D. (2001), ‘The Effects of Employee Satisfaction, Organizational Citizenship Behavior,and Turnover on Organizational Effectiveness: A Unit-Level, Longitudinal Study,’Personnel Psychology, 54, 101–114.

The International Journal of Human Resource Management 21

Dow

nloa

ded

by [

Uni

vers

ite L

aval

] at

08:

55 1

8 Ju

ly 2

012

Krishnan, S.K., and Singh, M.S. (2010), ‘Outcomes of Intention to Quit of Indian IT Professionals,’Human Resource Management, 49, 3, 421–437.

Lam, W., Chen, Z., and Takeuchi, N. (2009), ‘Perceived Human Resource Management Practicesand Intention to Leave of Employees: The Mediating Role of Organizational CitizenshipBehaviour in a Sino-Japanese Joint Venture,’ The International Journal of Human ResourceManagement, 20, 11, 2250–2270.

Lavelle, J.J. (2010), ‘What Motivates OCB? Insights from the Volunteerism Literature,’ Journal ofOrganizational Behavior, 31, 918–923.

Lee, T.W., Mitchell, T.R., Sablynski, C., Burton, J., and Holtom, B.C. (2004), ‘The Effects of JobEmbeddedness on Organizational Citizenship, Job Performance, Volitional Absences andVoluntary Turnover,’ Academy of Management Journal, 47, 711–722.

Lichtenstein, R., Alexander, J.A., McCarthy, J.F., and Wells, R. (2004), ‘Status Differences inCross-Functional Teams: Effects on Individual Member Participation, Job Satisfaction, andIntent to Quit,’ Journal of Health and Social Behavior, 3, 322–335.

MacKenzie, S., Podsakoff, P., and Ahearne, M. (1998), ‘Some Possible Antecedents andConsequences of In-Role and Extra-Role Salesperson Performance,’ Journal of Marketing,62, 87–98.

Meierhans, D., Rietmann, B., and Jonas, K. (2008), ‘Influence of Fair and Supportive LeadershipBehavior on Commitment and Organizational Citizenship Behaviour,’ Swiss Journal ofPsychology, 3, 131–141.

Miller, H., Katerberg, R., and Hulin, C. (1979), ‘Evaluation of the Mobley, Horner, andHollingsworth Model of Employee Turnover,’ Journal of Applied Psychology, 64, 509–517.

Mitchell, T.R., Holtom, B.C., Lee, T.W., Sablynski, C.J., and Erez, M. (2001), ‘Why People Stay:Using Job Embeddedness to Predict Voluntary Turnover,’ The Academy of ManagementJournal, 44, 6, 1102–1121.

Mobley, W. (1977), ‘Intermediate Linkages in the Relationship Between Job Satisfaction andEmployee Turnover,’ Journal of Applied Psychology, 62, 237–240.

Mobley, W., Horner, S., and Hollingsworth, A. (1978), ‘An Evaluation of Precursors of HospitalEmployee Turn-Over,’ Journal of Applied Psychology, 63, 408–414.

Morrison, E.W. (1994), ‘Organizational Citizenship Behavior as a Critical Link Between HRMPractices and Service Quality,’ Human Resource Management, 35, 4, 493–512.

Mossholder, K.W., Settoon, R.P., and Hanagan, S.C. (2005), ‘A Relational Perspective On Turnover:Examining Structural, Attitudinal, and Behavioural Predictors,’ Academy of ManagementJournal, 48, 4, 607–618.

Mowday, R.T., Koberg, C.S., and McArthur, A.W. (1984), ‘The Psychology of the WithdrawalProcess: A Cross-Validational Test of Mobley’s Intermediate Linkages Model of Turnover inTwo Samples,’ The Academy of Management Journal, 27, 1, 79–94.

Murphy, K.R. (1983), ‘Fooling Yourself with Cross-Validation: Single Sample Designs,’Personnel Psychology, 36, 111–118.

Ng, T.W.H., and Sorensen, K.L. (2008), ‘Toward a Further Understanding of the RelationshipsBetween Perceptions of Support and Work Attitudes,’ Group & Organization Management,3, 243–268.

Organ, D. (1988), Organizational Citizenship Behavior: The Good Soldier Syndrome,Lexington, MA: Lexington Books.

Organ, D. (1990), ‘The Motivational Basis of Organizational Citizenship Behavior,’ Research inOrganizational Behavior, 12, 43–72.

Organ, D.W., Podsakoff, P.M., and MacKenzie, S.B. (2006), Organizational Citizenship Behavior:Its Nature, Antecedents, and Consequences, Thousands Oaks, CA: Sage.

Paille, P. (2009), ‘Assessing Organizational Citizenship Behaviour in the French Context: Evidencefor the Four Dimensional Model,’ Journal of Psychology, 143, 2, 133–146.

Paille, P., and Grima, F. (2011), ‘Citizenship and Withdrawal in the Workplace: RelationshipBetween Organizational Citizenship Behaviour, Intention to Leave Current Job and Intention toLeave the Organization,’ Journal of Social Psychology, 151, 4, 478–493.

Peters, L., Jackofsky, E., and Salter, J. (1981), ‘Predicting Turnover: A Comparison of Part-Timeand Full-Time Employees,’ Journal of Occupational Behavior, 2, 89–98.

Podsakoff, P.M., and MacKenzie, S.B. (1994), ‘Organizational Citizenship Behaviors and Sales UnitEffectiveness,’ Journal of Marketing Research, 31, 351–363.

P. Paille22

Dow

nloa

ded

by [

Uni

vers

ite L

aval

] at

08:

55 1

8 Ju

ly 2

012

Podsakoff, P.M., and MacKenzie, S.B. (1997), ‘Impact of Organizational Citizenship Behavioron Organizational Performance: A Review and Suggestion for Future Research,’Human Performance, 10, 2, 133–151.

Podsakoff, P., MacKenzie, S., Paine, J., and Bachrach, D. (2000), ‘Organizational CitizenshipBehaviors: A Critical Review of the Theoretical and Empirical Literature and Suggestions forFuture Research,’ Journal of Management, 26, 3, 513–563.

Rosse, J.G. (1988), ‘Relations Among Lateness, Absence, and Turnover: Is There a Progression ofWithdrawal?’ Human Relations, 41, 517–531.

Sager, J.K., Griffeth, R.W., and Hom, P.W. (1998), ‘A Comparison of Structural ModelsRepresenting Turnover Cognitions,’ Journal of Vocational Behavior, 53, 254–273.

Sagie, A., Koslowsky, M., and Hamburger, Y.A. (2002), ‘Antecedents of Employee Lateness: AMultiple-Level Model,’ in Voluntary Employee Withdrawal and Inattendance. A CurrentPerspective, eds. M. Koslowsky and M. Krausz, New York: Kluwer Academic, pp. 1–20.

Schnake, M.E. (1991), ‘Organizational Citizenship: A Review, Proposed Model, and ResearchAgenda,’ Human Relations, 44, 735–759.

Schaninger, W.S., and Turnipssed, D.L. (2005), ‘The Workplace Social Exchange Network: ItsEffect on Organizational Citizenship Behavior, Contextual Performance, Job satisfaction, andIntent to Leave,’ in Handbook of Organizational Citizenship Behavior: A Review of‘Good Solder’ Activity in Organizations, ed. D.L. Turnipseed, New York: Nova Science,pp. 209–242.

Spector, P.E. (1987), ‘Method Variance as an Artefact in Self-Reported Affect and Perceptions atWork: Myth or Significant Problem?’ Journal of Applied Psychology, 86, 1, 114–121.

Thau, S., Bennett, R.J., Stahlberg, D., and Werner, J.M. (2004), ‘Why Should I Be Generous When IHave Valued and Accessible Alternatives? Alternative Exchange Partners and OCB,’ Journal ofOrganizational Behavior, 25, 607–626.

Tolich, M.B. (1993), ‘Alienating and Liberating Emotions at Work: Supermarket ClerksPerformance of Customer Service,’ Journal of Contemporary Ethnography, 22, 3, 361–381.

Turnipseed, D. (2002), ‘Are Good Soldiers Good? Exploring the Link Between OrganizationalCitizenship Behavior and Personal Ethics,’ Journal of Business Research, 1, 1–15.

Vandenberg, R., Lance, C., and Taylor, S. (2005), ‘A Latent Variable Approach to Rating Source ofEquivalence: Who Should Provide Ratings on Organizational Citizenship BehaviorDimensions?’ Handbook of Organizational Citizenship Behavior: A Review of ‘Good Solder’Activity in Organizations, ed. D.L. Turnipseed, New York: Nova Science, pp. 109–141.

Van Dyne, L., and Cummings, L. (1990), ‘Extra-Role Behaviors: The Need For Construct andDefinitional Clarity,’ Paper presented at the Academy of Management Annual Meeting,San Franscisco, CA.

Wermeling, L. (2009), ‘Social Work Retention Research: Three Major Concerns,’ Journal ofSociology, Social Work and Social Welfare, 3, 1, 1–21.

Yao, X., Lee, T.W., Mitchell, T.R., Burton, J.P., and Sablynski, C.J. (2004), ‘Job Embeddedness:Current Research and Future Directions,’ in Innovative Theory and Empirical Research onEmployee Turnover, eds. R. Griffeth and P. Hom, Greenwich, CT: Information Age,pp. 153–187.

The International Journal of Human Resource Management 23

Dow

nloa

ded

by [

Uni

vers

ite L

aval

] at

08:

55 1

8 Ju

ly 2

012