Development and validation of the Attitudes Towardsthe Homeless Questionnaire

H E Lester & H M Pattison

Context The homeless are a signi®cant group within

society, which is increasing in size. They11 have demon-

strably greater physical and mental health needs than the

housed, and yet often have dif®culty accessing primary

health care. Medical `reluctance' to look after homeless

people is increasingly suggested as part of the problem.

Medical education may have a role in ameliorating this.

Objectives This paper reports on the development and

validation of a questionnaire speci®cally developed to

measure medical students' attitudes towards the

homeless.

Method and results The Attitudes Towards the Home-

less Questionnaire, developed using the views of over

370 medical students, was shown to have a Pearson

test±retest reliability correlation coef®cient of 0á8 and a

Cronbach's alpha coef®cient of 0á74.

Conclusions The Attitudes Towards the Homeless

Questionnaire appears to be a valid and reliable

instrument, which can measure students' attitudes

towards the homeless. It could be a useful tool in

assessing the effectiveness of educational interventions.

Keywords *Attitude of health personnel; education,

medical, undergraduate; evaluation studies; *homeless

persons; questionnaires; *students, medical.

Medical Education 2000;34:266±268

Introduction

Medical education is a professional socialization

experience which involves the acquisition not only of

knowledge and skills but also of attitudes, values and a

sense of ethics. It has even been argued that medical

education has indeed a duty to supply society with

doctors who are willing to provide service equitably and

ef®ciently to all human beings, regardless of back-

ground or social standing.1

The homeless represent a group of people who are

frequently dismissed and stigmatized as troublesome

and unwanted patients, yet they are a signi®cant group

within society, which is increasing in size2 . Of®cial ®gures

for homelessness show that in England between 1984

and 1995, 1á42 million households were accepted by

local authorities as `statutorily homeless'.2

Extrapolating from the number of single homeless in

London, there are also tens of thousands of single

homeless in the United Kingdom at any one time. The

homeless also have demonstrably greater physical and

mental health needs than the housed3 yet they often

have dif®culty accessing primary health care. Practice

features, such as in¯exible appointment systems and

of®cious receptionists, and disincentives within `the

system', such as the lack of ®nancial inducements for

doctors, and workload implications have been suggested

as being the major barriers to providing primary health

care within mainstream services. However, medical

`reluctance' to register homeless people has been

increasingly suggested as a factor which explains low

registration rates especially among the single homeless.3

There has been little previously published work

aimed at measuring professional or public attitudes

towards the homeless. The `attitudes towards

homelessness inventory', developed in the United

States, is a validated 11-item Likert scale question-

naire.4 However, there appears to be no United

Kingdom equivalent, and no questionnaire which

speci®cally addresses the views on the homeless of

medical students or health professionals.

Development of the Attitudes Towardsthe Homeless Questionnaire

The lack of a suitable questionnaire led us to develop our

own instrument according to standard psychometric

Department of General Practice, The Medical School, University of

Birmingham, Birmingham, UK

Correspondence: H E Lester, Department of General Practice, The

Medical School, University of Birmingham, Edgbaston, Birmingham,

UK

Research papers

266 Ó Blackwell Science Ltd MEDICAL EDUCATION 2000;34:266±268

principles, which we subsequently used to evaluate the

effect of a 20-hour module on homelessness and health

offered to second-year medical students at Birmingham

Medical School. The original instrument contained 30

5-point Likert scale item statements, ranging from a

score of 1, where students reported that they strongly

agreed with the statement, to a score of 5, where they

reported that they strongly disagreed with the state-

ment. The content validity of the instrument was

increased by deriving all the items from verbatim

responses in previous in-depth interviews with general

practitioners in Birmingham about providing medical

care for homeless people. The item pool included value

judgements about the personal and societal causation of

homelessness, motivations for applying for medical

school and willingness to af®liate with the homeless.

Each item was coded so that a higher score re¯ected a

more positive response to the homeless. Half of the

original item statements were reversed.

During its development in 1996±97, the question-

naire was administered to over 200 medical students in

the ®rst, second and ®nal years at Birmingham Medical

School, and also to a cohort of 170 students within days

of receiving their A-level results, crucial to securing a

place at the Medical School. Students used their uni-

versity numbers as identi®cation for repeat question-

naires so that individual anonymity could be assured.

Students were therefore not pressurized into giving

`politically correct' answers.

Data analysis

Data were analysed using the statistical facilities of the

spreadsheet package Excel. Item correlation analysis

led to the removal of 10 statements that failed to

achieve signi®cance at the 5% level, resulting in a ®nal

20-item questionnaire (see Appendix) with a total

possible score of 100. The questionnaire required

3±5 minutes to complete.

The Pearson test±retest reliability correlation coef®-

cient of the 20-item version was 0á8. The Cronbach's

alpha coef®cient, testing the internal consistency of the

items, was good at 0á74.5

The health and homelessness module taken by 11

second-year medical students was evaluated using the

20-item version of the Attitudes Towards the Homeless

Questionnaire, which was completed by each of the

students on the ®rst and last day of the module. The

group's mean score before taking the module was 81á6.

The 20-item questionnaire was also distributed to two

other groups of 11 students who had chosen biological

science modules offered by other departments in the

Medical School, one module being about temperature

control and the other about pain management. The

mean scores of these two groups were 70á4 and 73á9.

Comparison of the scores of the three groups before the

modules started, using a paired t test, showed that the

scores of the homelessness module student group were

already demonstrably signi®cantly different from those

of some of their peers (t � 2á85, d.f. � 20, P < 0á01).

However, the two biological sciences student groups

were not signi®cantly different from each other. This

suggests that the questionnaire has good construct

validity.

After taking the module, the mean score of the stu-

dents in the health and homelessness group was 86á5.

Comparison of the questionnaire scores before and

after the module did not quite achieve statistical sig-

ni®cance (t � 1á9, d.f. � 20, P < 0á07), probably

because of the small number of students involved and

the `ceiling effect' of their pre-existing positive

attitudes; however, this suggests that the questionnaire

has good face validity.

Conclusions

Although the instrument requires further `road testing'

with different cohorts of students, and administration

before and after future health and homelessness

educational initiatives, the Attitudes Towards the

Homeless Questionnaire appears to be a valid and

reliable test, which can measure students' attitudes

towards the homeless. Instruments such as the Atti-

tudes Towards the Homeless Questionnaire could

perhaps also be used to select more socially motivated

medical students as an adjunct to traditional methods

of medical student selection. It could also be used to

measure the effectiveness of educational interventions

designed to improve attitudes towards homelessness at

all levels of medical education.

References

1 Knight JA. Doctor to be. Coping with the Trial and Triumphs

of Medical School. New York: Appleton Century Crofts; 1981.

2 Department of Employment. Information Bulletin.3 London:

DoE; 1996.

3 Burrows R, Pleace N, Quilgars D. Homelessness and Social Pol-

icy. London: Routledge; 1997.

4 Kingree JB, Daves WF. Preliminary validation of the attitudes

toward homelessness inventory. J Commun Psychol

1997;25:265±88.

5 Cronbach LJ. Coef®cient alpha and the internal structure of

tests. Psychometrika 1951;16:297±334.

Received 21 September 1998; editorial comments to authors

11 February 1999; accepted for publication 17 March 1999

Ó Blackwell Science Ltd MEDICAL EDUCATION 2000;34:266±268

Attitudes Towards the Homeless Questionnaire · H E Lester & H M Pattison 267

Appendix

Table 1 Attitudes Towards the Homeless Questionnaire

AGE

SEX

We are interested in ®nding out about medical students' and doctors' attitudes towards the homeless.

On the questionnaire, please circle the number that most closely represents your views.

There are no right and wrong answers to these questions. You do not need to spend long on each statement ± often your ®rst response is

the most accurate.

1 = strongly agree

2 = agree

3 = neither agree nor disagree

4 = disagree

5 = strongly disagree

1. Homeless people do not choose to be homeless 1 2 3 4 5

2. Nearly all homeless people are drug addicts 1 2 3 4 5

3. You only need to learn about homelessness if you want to be a GP 1 2 3 4 5

4. Homeless people are victims 1 2 3 4 5

5. Homeless people are rude 1 2 3 4 5

6. Doctors have a duty to care for the homeless 1 2 3 4 5

7. Homeless people are aggressive 1 2 3 4 5

8. Homelessness is a major problem in our society 1 2 3 4 5

9. Homelessness is a self in¯icted state 1 2 3 4 5

10. Homelessness is not a health issue 1 2 3 4 5

11. The NHS should not waste its resources on the homeless 1 2 3 4 5

12. I entered medicine because I want to help those in need 1 2 3 4 5

13. People make themselves homeless to get a better house 1 2 3 4 5

14. No one in this country has to `sleep rough' 1 2 3 4 5

15. The State should spend more money on providing housing 1 2 3 4 5

16. Alcoholism is a personal weakness 1 2 3 4 5

17. I entered medicine because I was good at science 1 2 3 4 5

18. Homelessness is not a signi®cant problem in the UK 1 2 3 4 5

19. Doctors should address physical not social problems 1 2 3 4 5

20. The State should spend more money on the care of the homeless 1 2 3 4 5

Attitudes Towards the Homeless Questionnaire · H E Lester & H M Pattison268

Ó Blackwell Science Ltd MEDICAL EDUCATION 2000;34:266±268