IlIV PRIMARY PRFVENTION AND VERBAL ATTITUDES AND BEH/\VIOURS A
STUDY OF TWO INTERVENTI01\JS

by
Maria A Walsh
B Ed., Memorial University of NeVvfoundland, 1990

THESIS SUBMITTED IN PARTIAL FULFILLMENT OF
THE REQl 1IRE~1E7\JTS I OR THE DEGREE OF
MASTER OF LDUC ATlON
m

COL,NSELLING

c Maria Walsh, 1998
THE UNIVERSITY OF NORTHERN BRITISH COLUMBIA
Pnnce George, British Columbia
April, l 998

All rights reserved. This work may be reproduced in whole or in par1 , by photocopy or other

means, Wlthout the permission of the author.

..

II

ABSTRACT

In an experimental process, three groups were tested pre and post
intervention to determine change in attitudes and behaviours. Two
interventions were designed, using Social Cognitive Theory as the
framework Intervention I, was an affective intervention subjects to
the teaching of a young heterosexual woman who was I IIV positive.
Intervention 2, involved sub.1ccts planning a prevention program for
members of their population. The sample group consisted of young
adults, ages 18 to 25. attending UNBC. The sample group was divided
into three study groups; Control, N
and Intervention Group 2. N

= 9:

Intervention Group I , N

=

IO:

12 The questionnaire package used to

measure change consisted of a demographic survey (only used in pre test),
an attitude and a behaviour questionnaire Results of ANO VA showed
no statistically significant difference among the three groups at pre test
for attitude and behaviour. However, an ANOV A done on gains (pre-post)
reflected statistically significant difference between control group and
Intervention l

Results indicate that an affective intervention involving a

person living with AIDS interacts with participants can be said to be
effective in promoting some behaviour change. More research needs to be
done to discover the behaviours most impacted by the mtcrvention
Open ended questions showed several different responses at the post test

111

le\'el particularly in the area of concern for self and others, increase in
detailed information. acknowledgement that abstinence is a viable option.
and acknowledgement that one could not identify a person as HIV positive.
Both interventions were c:een as viable for prevention programs, particularly
as they did not involve significant considerations and were very basic
and could be run in many different venues among many different groups.

IV

TABLE OF CONTENTS
ABSTRACT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

..
11

TABLE OF CONTENTS ..................... . . . . . . . . . . . . . .

1v

ACKNOWLEDGEMENTS . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . .

v1

..

DEDICATION . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . vu
CHAPTER ONE
rNTRODUCTION
Statement of Prohlem. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . l
Significance of Problem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
Explanation of Terms . . . . . . . . . . . . . . . . . . . . .
. ........ 7
Sum1nary. . . . . . . . . . . . . . . . . . . . . . . .................... 8
CHAPTER 2
REVIEW OF THE LITERATURE...... . . . . . . . . . . . . . . . . . . . . . . 9

I-iistory. . . . . . . . . . . . . . . . . . . . . . . . . . . . ................. 9
Epidemiological Information. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
Paths of Transmission.
. . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
HIV Primary Prevention Research. . . . . . . . . . . . . . . . . . . . . . . . 15
Theory and HIV Research. . . .
. . . . . . . . . . . . . . . . . 16
Factors and Barriers Affecting I IIV Preventive
Attitudes and Behaviours
........ __ . . . . . . . . . . . . . 25
Modification and Promotion of HIV Preventive
Attitudes and Behaviours. . . . . . . . . . . . . . . . . . . . . . . . . . 3 I
Challenges and Solutions in H1V Primary Prevention ........... 33
Sexual Behaviour _
. . . . . . . . . . . . . . . . . . . . 34
Cultural/Group Considerations
. . .. . . . . ......... 36
Denial and Underestimation of Risk . . . . . . . . . . . . . . . . 38
Research Hypotheses ................................. 40
Summary................ . ........ .. ................ 42

CHAPTER 3
METHODOLOGY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 43
Subjects. . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . 43
Setting. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
44
Research Instnunents
. . . . . . . . . . . . . . . . . . . . . 44
Demographic Questionnaire. . . . . . . . . . . . . . . . . . . . . . . . 45

V

Attitude Questionnaire
. . . . . . . . . . . . . . . . . . . . . . . . . 45
Behaviour Questionnaire .......................... 47
............ . .......... .. . . ............ . . 48
Procedure
Data Analysis . . . . . . . . ............................. 53
Sun1n1ary. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .............. 57
CH~PTER4
RESULTS

. . . . . . . . . . . . . . . . . . . . . . . . . 59

Discussion of Demographic Data. . . . . . . . . . . . . . . . . . . . . . . . . 59
Presentation of Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
Item Analysis
. . . . . . . . . . . . . . . . . . . . ... 63
Close Ended Responses . . . . . . . . . . . . . . . . . . . . . . 65
Open Ended Responses
. . . . . . . . .. . . ....... 67
Attttude Questionnaire . . . . . . . . . . . . . . . .. . . . . 68
Behaviour Questionnaire . . . . . . . .. . . . . . . . . . . . 69
Post Test Add1t1on ......................... 71
Discussion of Results. . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . 73
Surnmary. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ........... 75
CHAPTER 5
DISCUSSION

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77

General Discussion . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . 77
Hypothesis 1. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
Hypothesis 2. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
Limitations. .. . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . 81
Suggestions for Fur1her Research . . . . . . . . . . . . . . . . . . . . . . 86
Conclusion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
LIST OF REFERENCES

.................................... 91

APPENDIX A. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102

APPENDIX B .............................................. 104
APPENDIX C. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108
APPENDJXD ........................................... 115

VI

ACKNOWLEDGEMENTS
There are many people to whom I owe much gratitude for their help and
support over the past tour years Thank you, firc;t, to my co-supervisors
Ron, the lessons you taught are priceless Thank you so much for your patience
and for sharing your expertise with me.
Peter, your way of gently educating and your commitment to your students makes you
a very special teacher Who knew I would come to be fascinated by statistical
processes?!
Thank you, Annette, my third committee member, for your willingness to be on
my committee and to give me positive and valuable feedback l would also like to
express my appreciation to Glen Schmidt for his willingness to play the role of external
examiner Your input was invaluable. Also, a thank you to John Curry, who helped
make my defence a positive, comfortable, and very enjoyable experience
Thank you to Jan, for having the courage and canng to share your very special
experiences with a group of university students. You were tnily inspiring.
A heartfelt thanks to all friends and family who supported me and cheered me
on. Lissa, sister, we went through this together - words can't express what an
important part of my life you are. Kathy B , my cheerleader, every time I turned
around - there you were. Heather. you showed me those parts of academia of which T
want to be part, you are truly an inspired teacher and student. Thanks to my sisters,
Kelly, Coleen and Krista - you have all helped me keep going. Thanks, Coleen, for
getting that wonderful information for me, and Krista, for the chats on the ICQ
Counsellor Leonard, your sense of humour helped me keep perspective Thank you to
Barb for your friendship, support and all of your help coding It is a good thing that we
laughed as much as we worked Thanks Ray for your help in the testing and data
collection process Thanks to everyone in "Bits and Pieces" for being so interested and
for helping me maintain some level of sanity and connection to the world outside the
university. Thanks Catherine for the instigation to keep going! And, finally, to Sean DITTO!!

Vil

DEDICATION

This work is dedicated to my parents, Pat and Alfreda Walsh :
You have 11ever doubted that your girls cou]d do anything they wished.
Why shouJd {?
And to my dear friend Ross:

You are the inspiration for this work
Your cow·age never ceases to amaze me!

CHAPTER ONE
lNTRODUCTION

In Chapter One, information about the fonnulation of the 5ludy is presented

along with brief outlines of some of the relevant literature. Lssential elements to the
study are explored This Chapter also includes a statement of perceived significance of
the study and an explanation of any terms necessary to the research that may have
needed further clarification.

Statement of Problem
The Human Immunodeficiency Virus OJJV) and the resultant syndrome,

Acquired Immune Deficiency Syndrome (AIDS) has become one of the most profound
concerns of the latter 20th century. According to a report from Health Canada ( l 997a):
As of June 30. 1997. there had been a cumulative total of J 5, IO 1 reported
cases of AIDS in Canada since the beginning of the epidemic... After
adjusting for under reporting and delayed reporting, we estimate that, by
the end of 1996. there had actually been about 20,000 AIDS cases in Canada
since the epidemic began ( p. l ).

Health Canada (1996) also presented the results of surveillance of HIV contraction since
198~ From 1985 to 1995, 36,613 cases of HIV were detected in Canada. With the

effects of this vin1s being felt globally. the rush to design and implement usefol
prevention programs grew To date. there is no cure for HIV, no cure for the resultant

syndrome and no vaccine to prevent contraction.

2

While millions of dollars were being poured into AIDS research, the number of
people living with l IIV continued to grow, often to epidemic proportions Methods of
lessening the risk of contraction of HIV did exist and were spread over the media for
several years /imbardo and Lieppe ( 199 l) \-varncd that information was not being acted
upon, despite the media blitz:
At the level of national media, the blitz of 1 V and radio messages
designed to persuade sexually active people to use condoms as precaution
against the AIDS virus has enjoyed only mixed success (p. 129).
Generally, prevention eftbrts were seen as meeting with similar dubious outcomes
Behaviour was not being changed (Fisher, Fisher, \1isovich. Kimble, & Malloy, 1996 ).
Keeling ( 1991) reported that college students "in general, have very good to excellent
levels of knowledge" (p 5 I). Ho~ever, he also asserted that this knowledge had not
decreased risky behaviours. Knowledge and positive attitudes were not always
precursors to increased safer sex activity, attitudes did not, necessarily, translate into
behaviour (Ajzen & Fishbein, 1980). Theorists attempted to define the reason for this
lack of behavioural change Researchers stated that those who developed intervention~
have not accorded enough importance to cultural influences Dancy ' s ( 1991) study on
developing ethnically sensitive and gender-specific questionnaires, stressed the
importance of this information to group appropriate intervention development. Other
authors have claimed that the correct infom1ation is not being received. understood
and/or accepted by the larger population Numerous studies (Bryan. Aiken, & \\ est,
1997, Rothspan & Read, 1996, van dcr Plight & Richard l 994~\Veinstein, 1989)

3

presented explorations of the concepts of perceived susceptibility and invulnerability and
implications of these concepts for behaviour change. In much of the research, the belief
that HIV contraction only occurred in homosexual men was put forward as partial
reason for lack of behaviour change. If such beliefs still exist. there are many men and
women who may not have been practising safer sex in the mistaken belief that they are
safe The Canadian Public Health Association ( 1991) defined primary prevention as the

only way to fight the spread of HIV and stressed that every possible intervention and
means of education must be considered and tested.
Kanfer and Goldstein ( 1991) defined prevention as the attempt "to intervene
before deficiencies would lead to symptoms of frank pathology" (p . 74 ). While this
definition was used in reference to cJmical perspective work, the relevance to a decrease

in HIV contraction was clear. Indeed, the concern was to prevent behaviours
(deficiencies) before contraction (pathology) occurred In this study, prevention or
prevention programs referred to a whole attempt to lessen the possibility of HIV
contraction, while interventions or preventive strategies were considered parts of that
whole
The role of preventive strategies is to promote positive attitudes and behaviour
toward preventive lifestyles. For the purposes of this study, attitudes were defined as the
psychological tendencies to view particular entities either positively or negatively (Eagly
& Chaiken, 1993) People evaluate objects and situations and from this evaluation

formulate a favourable or unfavourable attitude Attitudes are divided into three
subgroups, COGNITIONS - expressions of beliefs; AFFECT - expressions of feelings ,

4

CON ATI ON - expressions of intentions to behave in a given way (Krebs & Schmidt,
1993) These three aspects of attitude were considered in the development and

administration of the questionnaire package and in the interventions used in this study
Wolman ( 1989) defined behaviour5 as verbal expressions of interactions with the social
environment Verbal representations of both attitudes and behaviours related to l IIV
contraction were investigated.
For the current studv. the researcher considered the attitudes and behaviours of
young adults Two interventions were designed in an attempt to increase positive
attitudes and behaviours (definition of attitudes and behaviours can be found on page 13)
in young adults ages, 18 through 25 , who attended the University of Northern British
Columbia (UNBC). O'Leary, Goodhart, Jemmot, & Beecher ( 1992) expressed that
interventions were essential for young American adults due to the continuance of high
risk behaviour among this group. The decision to work with this age group also
evolved from theories of attitude flexibility and change In these theories. attitudes were
reported to become more persistent from an early point of adulthood (Krebs & Schmidt,
J993) . A study of change was presumed, by this researcher, to be more successful with

young adults than with older adults for whom persistence was increasmg.
The first intervention in this study was an affective inten ention. Interact ion with
a young heterosexual woman who had contracted HIV through heterosexual intercourse
was hypothesized by this author to aid students in associating with the tdea that the)
were, indeed, at risk The imparting of lived experience by the presenter would promote
a change in the evaluation of the attitude/bchav1our ob_1ect The potential for

5

observational learning was proposed to be strong in models with whom people felt they
could associate (Bandura, 1986).
The second intervention involved students in the process of dccision~making
about. and planning for, a program specificaJly designed to promote positive attitudes
and behaviours in post secondary students of their own age group. The second
intervention was designed to capitalize on this process, not the teaching aspect but the
process of planning the program. The researcher postulated that subjects would
experience an increased personal involvement with the attitude/behaviour object. This
personal involvement would increase self-efficacy (Bandura, 1986) and feelings of
ownership of the program. The result of an increase m self-etlicacy and ownership was
expected to lead to positive change in attitude and behaviour and/or maintenance of
positive attitudes an<l behaviours
The use of an established theoretical framework is a very important aspect of
intervention design (Fisher, Fisher, & Rye, 1995, Mantell, DiVittis. & Auerbach, 1997).
Kowalewski, Henson & Longshore ( 1997) noted the lack of interventions developed
from specific theories of social learning. The Health Belief Model had been used rn many
research endeavours, but was not reported to be as effective as hoped (Wulfort & Wan,
1995) Such theories as the Theory of Reasoned Action and the Social Cognitive

Theory were not well represented in the literature on HIV prevention. The researcher
speculated that this may have been due lo the lack of reporting of the theory behmd a
particular piece of research (denying the reader valuable information). Another possible
reason was the fact that interventions were developed based on procedures without

6

theory . Many interventions v.ere designed according to established steps and processes,
but not grounded in theory. A theoretical base that could help tho~c who were designing
interventions to address social, cultural, affective and cognitive issues relevant to Ill V
prevention was also needed (Burnette, I996: J Ierlocher. I Iofl~ & DeCarlo, I 996~Truax,
1994 ).

Significance of the Problem
Contraction of HIV is a major health concern in th1~ era lhere is no doubt that
the impact i~ being felt globally. Due to the lack of a cure, 5trateg1es have focused on
prevention as the key to control1ing the spread of HIV (Edgar, Fitzpatrick, & freemuth,
t 992). Development of effective prevention programs has been presented as requiring

(I) design of interventions geared toward particular groups and (2) proof of efficacy in
promotion of attitude and behaviour change through evaluation. Interventions proven
effective in certain populations can then become part of a larger prevention program that
would be tested for appropriate interaction between interventions (Mantell ct al. , 1997).
In this study, the interventions were designed to address HIV issues relevant to young
adults 18 - 25 . Efficacy of these interventions, significant change in attitude and
behaviour, would have allowed further exploration of the impact of similar interventions
on the same group with the eventual design oflarger prevention programs with the
effective interventions as major components.
Attitudes about HIV play an important role in prevention King, Beazley,
Warren, Hankins, Robertson, and Radford ( 1989) stressed the importance of attitudes in
HIV prevention when they stated that it was essential to provide "... opportunities for

7

young people to develop positive attitudes predisposing them to safer, health promoting
behaviour"(p I42) It was proposed that the two interventions in this study would aid
students in developing more positive attitudes toward all forms of prevention.
While the role of attitudes in prevention is important. the role of behaviour
change should be the ultimate goal of intervention development and evaluation. It was
hypothesized that interventions combining presentation of accurate information, along
with interventions designed to promote 111creased self efficacy through modelling or
ownership and through goal setting, would effect positive behaviour change.

Explanation of Term~
Most of the terms essential to this research were either self-explanatory or
explained in the "Statement of the Problem'' section of this chapter. A few terms used
throughout this work may need clarification for the reader. In many areas references
were made to positive attitudes and behaviours

For the purposes of the study, those

attitudes and behaviours defined as leading to increase in potential for contraction were
seen as negative (see Chapter 3 for attitudes and behaviours of concern). Those
attitudes and behaviours considered preventive in nature, such as increased use of
condoms during sexual intercourse, were considered positive,
Throughout the study, the tem1s 'preventive strategies' and 'interventions' \\ere
used interchangeably. These terms referred to any part of a prevention program. The
researcher did not define one test or a single interaction with a target group as a
prevention program. A program was interpreted as an integration of interventions
designed to help targeted populations in preventing contraction of I IIV

The expression

8

of sex denoted sexual intercourse

Summary
In this chapter, the researcher outlined the purposes of the 5ludy, including some
relevant literature leading to the development of the study. Significance of the problem
is presented and any terms that needed fi1rther explanation arc defined . The following
chapters represent the progression or the study from investigation of the relevant
literature through investigation of results and possible impact In Chapter 2, information
from various studies and writings m the areas relevant to design, implementation and
evaJuation of I IIV interventions 1s presented. rn Chapter 3, the reader is provided with a
detailed account of the methodological considerations for this study from subject
selection to proposal of types of stat1st1cal analyses to be used . The results of analyses
are explored in Chapter 4 with discussion of significance of these re!>ults In Chapter S.
the researcher presents an interpretation of the results in relation to the research
questions Relevance of the thesis to HIV prevention is discussed as well as limitations
in this research. In the final section of the fifth chapter, the researcher presents
suggestions for further related work based on the outcomes of this study and continued
research into relevant literature.

9

CIIAPTER 2
R[•VIEW OF THE LITERATURE

fn Chapter 2, a compilation of the findings from an in depth investigation of
literature relevant to HT\' prevention 1s presented ·1he review begms with a bnef
exploration of historical and epidemiological infom1ation concerning AIDS/HIV. This
exploration includes the origin of HIV, cu1Tent cpidemiolog1cal details of J 11 VIAIDS
presence in Canada and infonnation about contraction and transmission . The second
section of the review involves presentation of information around IIIV prevention
research including exploration of the use or lack of use of theory in HIV prevent10n,
overviews of studies that attempted to define factors related to risk and reduction of risk
and presentation of information gathered from studies that involved the development and
evaluation of interventions designed to modify attitudes and behaviours. The thtrd
section reflects various chaJlenges to primary prevention strategies and ways m which
researchers have attempted to deal effectively with these challenges In the final section
of the review, the research hypotheses are presented A brief summary highlights the
major themes m the information.
History
Specific proof of the origins of the Human Immunodeficiency Virus does not
exist. However, an accepted belief is that HIV is related to the Simian or Green
Monkey. "The virus first affected humans m central Africa where the monkeys were
hunted for food and often skinned and eaten raw. The monkeys also hite, increasing

10

the potential for infection .. (Learning Together About HIV. 1994, p. 4). Although
not name<l at the time, AIDS was described hy the Centre for Disease Control (CDC)
in January, 198 l. The development of a rare form of cancer. Karposi's Sarcoma
(KS), and a rare form of pneumonia calle<l Pneumocystis Carinii Pneumonia (PCP)
in young men focused attention on a cohort of illnesses that should not have heen
present in these men. The illnesses were seen only with severely immunodepressc<l
people. At that time. no specific cause had heen identified for the development of
these illnesses. The one common factor , however, was an underlying immune
system failure. Acquired Immune Deficiency (AIDS) was eventually used to describe
the cohort of symptoms including; weight loss, fatigue, PCP, KS, and premature
death. This definition of symptoms occurred in I982 when "the Centers for Disease
Control (CDC) in Atlanta, Georgia officially classified the conditions as Acquired
Immune Deficiency Syndrome, by which name it is still widely described" (Watney,
1994, p. 76). The causal factor of AIDS was fir'it isolated in 1983 hy Dr. Luc

Montagnier of the Pasteur Institute in Paris. However, Dr. Montagnier had difficulty
convincing his colleagues that this was the virus. It was not until May, 1984 that
HIV was designated. Dr. Robert Gallo, a researcher for the National Cancer Institute
(NCI), also isolated the virus and had less difficulty convincing the community that it

was the virus that led to the expression of AIDS symplomology.

11

Epidemiological Tnformat.iID.l
At the end of 1982, I l cases of A IDS had hecn reported in Canada. The

numher rose to 49 in 1983, tripled to 165 in 1984 and then more than doubled to 415
hy the end of 1985 (Learning Together Ab.uut HIV, 1994). In 1996. Health Canada

released a surveillance report of incidences of HIV in Canada. That report showed
the progres\ion of the virus from 1985 to the end of 1994. Between 1985 and 1994
lhere were 33,520 ca<;e, of HIV reported. f n 1995 there were 3093 new cases. This

hrought the total of people infecte<l up to 33,613 hy the end of 1995. Other
information provided in this report included:
1.

The total number of HIV ca~s in British Columhia in this time period
was 8819, more than 20% of the national number.

2.

The national number for adolescents. aged 15 to 19, was 510.

3.

The national numher of ca~es for young adults, 20 to 29, was 9992.

4.

The total number of cases of contraction through heterosexual contact
was 1376.

In an HIV/AIDS Epi Update called AIDS and HlV in Canada. "the Bureau of
HIV I AIDS and STD estimate(d) that as of the end of 1996, a cumulative total of

50 000-54 000 Canadians had been infected with HIV since the onset of the
epidemic ..... "(Health Canada, 1997, p. 2). The Bureau of HIV/AIDS and STD is a
branch of Health Canada at the Lahoratory Centre for Disease Control.

12

Paths of Transmission. In Western Society, AIDS wa~ defined as a homo~exual
disease. This generalization has continued although drug user~. hemophiltacs. and
prostitutes were added to the li<it of those who contracted. A.s expressed in the I994
resource, Learning Together About HIV,
such a generalization could lead to drastic results because numerous
sexual contacts was no longer the common denominator, nor was
homosexual contact. What was emerging was a pattern of infection
hy a pathogen trnnsmitted through the exchange of ho<ly fluids.

especially blood and semen (p. 4).
A definition of high risk behaviour did not focu~ on juc;t sexual contact. hut moved to
the exchange of particular body fluids. The refinement of the causes of HIV did not

seem to have become a reality for those who saw themselves as separate from the
early established high-risk groups. This meant that the heterosexual community
continued to risk exposure to HIV (Yarber, 1995).
Health Canada ( 1997h) deterrnined that Canadians are still engaging in sex
with multiple partners, even if this is a result of serial monogamy. Many men and
women engaging in sexual activity still do not report using condoms consistently, nor
do a percentage of intravenous drug users (lDU's) and men and women in sex-trade.
The Bureau of HIV/AIDS and STD's reported that
The number of new HIV infections that occurred in Canada in l 996

13

is estimated to he between 3,000 and 5,000, which is higher than the
estimated average number of 2.500-3,000 per year for the period
1989-1994. The majority of this im;reasc appears to he occurring among

injection drug users and young gay men . However. availahlc data also
suggest increasing HIV infections among non-lDU heterosexuals
(especially women) anti Aboriginal peoples (p. 2).
There is no cure for AIDS , and contraction of HIV is usually a devastating
experience. That devastation increases with the onset of the symptoms called AIDS .
According to Edgar ct al. ( 1992) prevention is the most appropriate way to deal with
this pandemic until c;;uch a time as the cure is found . This group of researchers
suggested that, in designing effective prevention strategies, it is essential to recognize
two very important aspects of human behaviour:
Prevention through appropriate hehaviour was and still is, the hest
weapon available to fight further spread of HIV infection. However,
individuals take necessary actions to prevent a disease such as AI OS
only when (a) they are properly informed and (h) they feel motivated
to respond to the information they p<>\sess (p. xi).
The responsihility for the development of preventive strategics lies with those who
function in the realm of the Social Sciences, community organizations. and
communities/people themselves (Ratzan. 1993 ; Orians, Lichow & Branch, 1995) .

14

These three groups. with some support hy the gcl\'crnmcnt. have become the active
members to provide those strategies necessary to decrease the numhcrs of people
contracting the virus .
HIV is found in all hody fluids to a greater or lesser extent. In some fluids
such as tears the \ irus is found in amounts that would not suggest much potential for
transm1,,ion. However, in semen. vaginal flui<l and hlood. the concentrations arc
high and thus there is risk for transm1,~1on (Learning Together About HIV, 1994) .
Hales ( 1994) outlined those situations and hehaviours that put pcopJc at risk for HIV
infection . The first wa\ ,exual activity. Anal and vaginal intercourse without the use
of a condom and/or non-oxyl 9 increa\e<l the potential for contraction . ff a pef\on
had already contracted an STD. chlamydia for example, risk is further increased. A
second set of nsky behaviours revolved around intravenous drug use . Intravenous
drug users (IDU's) may share needles using them to inject drugs directJy into the
bloodstream. If one of the IDU ·s \haring the needle was HIV positive the virus
would he injected into the system along with the drug. IDU 's who then have sex
without using condoms faced the risk of trnnsferring the virus to their sexual partner,
" ... . sex with infected IV-drug users has heen the numher one cause of HIV infection
in women"(p. 459). Blood products tainted with HIV were also identified as a point
of transmission, however, this occurrence is rare now, in Canada and the United
States, as a result of availahle screening methods. Mothers can pass the virus to their

J5

habies and people have contracted through accidental contact with infected hlood
and/or hody fluids through such things as needle sticks or heing splashed. Most of
the transmissions have occurred through sexual contact and needle sharing.
HIV Primary Prevention Research
It was acknowledged that the effort to develop programs to prevent spread of

HIV needs to he a shared effort hy many contributors. Public health workers, social
and behavioural scienti~ts. community groups, at-risk populations, government
agencies and program development specialists need to integrate their efforts to provide
the hest programs possible (Needle, Brown, Coyle, & Weissman, 1994; Snider &

Satcher, 1997: Valdeserri et al., 1995). There is increasing interest in and argument
for the role of social and behavioural science research in development of interventions
and programs. In investigating the literature, the researcher found that studies done
in the area of HIV prevention revolved around estahlishment and/or proof of
theoretical frameworks useful in development of preventive interventions,
investigation and exploration of contrihuting factors to HIV preventive attitudes and
behaviours and design, implementation, and evaluation of programs and interventions
to promote HIV preventive attitudes and behaviours in individuals and groups.
Acknowledgement is now being given to the role of scientists in explaining behaviour,
providing tested strategies for behaviour change and exploring the personal and social
challenges that are factors in attitudes and high risk behaviours. ( Burnette, 1996;

16

Clay, 1996; Holtgrave, Doll & Harrison. 1994 ).
Theory and I IIV Research. Herlocher et al. ( I 996) asserted that all prevention

practices followed at least informal theory ; knowledge of populations, and
determination of needs of specific populations are examples of informal theoretical
factors. They stressed that formal theories could also he very helpful in the provision
of explanations of behaviour and/or potential goals for preventive strategies and
programs. Several theories, many of which were social cognitive, were used in the
development of general health preventive programs or research studies and also in
HIV prevention specifically. Mantell et al . ( 1997) presented several of the theories
that had been used in interventions to promote HIV preventive behaviour as well as
the researchers responsible for their development. These theories included: (a) the
Theory of Reasoned Action - introduced hy Fisher and Ajzen ~ (h) the Health Belief
Model - developed by Rockenstock; (c) Social Cognitive Theory - modified by
Bandura; (d) the AIDS Risk Reduction Model - presented hy Catania, Kegefes and

Coates.
According to the Theory of Reasoned Action (TRA) intent to execute

behaviours is mediated by a person's attitude toward the particular behaviour, the
outcome expectancies of performing the behaviour and/or the normative heliefs
developed through perceptions of expectations of significant others with regards to the
performance of the behaviour (Baker, Morrison, Carter & Verdon 1 1996). Behaviour,

17

are then predicted from intentions to behave. In application of TRA to HIV
prevention, Fisher et al. ( 1995) stated that "an individual's attitude toward
performance of a particular AIDS preventive behaviour is a function of the
individual' s helief~ about the consequences of performing the hehaviour, multiplied

by his or her evaluations of these consequences" (p. 256). They added that the ahovc
concept was also influenced hy the person's perception of ~oc1al support and their
motivation to behave as expected. A,1,en and r. ,shbein ( 1980) pointed to the
importance of exploration and identification of heJiefs that a group hold~ ahout the
consequences of certain preventive behaviours and of determining those who would be
considered influential in the group in order to address the particular population.
TRA has been used in attempts to understand risk reduction behaviour and
social influences impacting this behaviour. Jn a study done on young gay men, Ross
and McLaws (1992) established predictors of condom use. The result~ showed that
opinions of others, significant in these mens' lives, were greater predictors of
intentions to use condoms than attitudes about condom use. The idea that intentions
were predictors of behaviour was also supported. They stressed the importance of
considering the influence of significant others when attempting to promote preventive
behaviours.
Fisher et al. (1995) investigated determinants of HIV preventive behaviours.
This study involved gay men, heterosexual university students and heten>\exual htgh

18

school students. All of these populations were defined as being at risk. A strong
correlation between expected consequences of the acts multiplied with sodal and
personal evaluations of these consequences and attitudes toward preventive behaviours
was found. They al. o discovered a relationship between intent to act and attitudes and

norms; this intent to act mediated behaviour. The researchers suggested that this
study proved TRA could he used to structure HIV preventive interventions. An added
element to intervention design based on TRA was suggested by Baker et al. ( 1996).
They reported that previous experience with condoms and their use influenced
intention to use the condoms. Therefore> it was necessary to investigate mechanisms
to explain the relationship between previous experience\ with condoms and condom
use.
The Health Belief Model (HBM) has been used hy researchers in an attempt to

explain health behaviour and efforts to change hehaviour by those seeking to improve
health status (Thurman & Franklin, 1990). Rosenstock, Strecher and Becker ( J988)
reported that use of HBM has Jed to significant results in research, although they
tempered this with the conclusion that the variance in behaviour explained hy HBM
was lower than expected. HBM was "designed to explain health actions in terms of
health related heHefs and motivations" (Petosa & Jackson, 1991 ). The perception of
efficacy of preventive behaviours was compared to perceived barriers to committing
to these preventive behaviours. In the case of HIV prevention, the behaviours would

)9

have included safer sex practices and the harriers would have included psychological,
social and or physical factors. According to Mantell et al. ( 1997) and Rosenstock et
a1. ( 1988), health preventive hehaviour was a result of the occurrence of three factors.
The first factor was the perception of threat. There must he enough concern ahout the
threat of the illness to make this threat relevant. The second factor involved
perceived susceptibility: "Perceived susceptihility refers to the likelihood of
experiencing personal harm if no action is taken ... "(Weinstein, 1989). The final
factor involved the weighing of perceived etlicacy of the preventive hehaviour and
barriers to this behaviour. Rosenstock et al. ( 1988) stated this factor involved "the
belief that following a particular health recommendation would he beneficial in
reducing the perceived threat and at a subjectively-acceptable cost." (p. J 77). A
person would conclude that the benefit of reduction of the risk of illness would
outweigh the energy spent overcoming barriers. The belief a person had in their
ability to carry out the necessary behaviours (self efficacy) was added to the HBM in
the 1980's (Mantell et al., t 997).

HBM has been used in attempting to identify contributing factor~ to the
adoption of preventive actions as we11 as to develop interventions. While Thurman
and Franklin (J 990) reported that the HBM was not sufficient to predict preventive
behaviour, they did find support for one of the primary concepts of HBM. This
concept, as cUscusse<l in the previous paragraph, was the perception of relevance of

20

threat to people. The subjects were studied regarding their reactions to the health
threat of HIV. It was reported that ~uhjects "were reluctant to change their sexual
behaviour unless the threat of infection wa~ personalized" (p. 179). Weinstein ( 1989)
also supported the importance of another m~jor factor included in HBM perceived
susceptibility. Weinstein expressed the opinion that many perceptions of peVional
susceptibility were not realistic and that many people ,howc<l a consistent trend
toward optimistic bias; they perceived persona] threat to be less than that of others in
their group. Considering this conclusion, any intervention that dealt with attempts to
promote preventive attitudes would need to involve elements to address optimistic hias
and so attempt to increase perception of threat and per')onaJ \usceptihility.
Petosa and Jackson (1991) used HBM to predict safer sex intentions of
adolescents. The results suggested that programs devoted to promotion of safer sex
intention needed to focus on health related motivations. The researchers discovered
an interesting outcome in this study. While the HBM effectively predicted intentions
in younger students it was not effective in predicting the same for older adolescents.
They concluded that for older adolescents more contributing factors influence
intentions than were specifically Jaid out in the HBM, such as self-presentation, need
for acceptance and other social influences.
Eisen and Zellman ( 1984) found a connection between health beliefs outlined
by HBM and knowledge about sexuality and contraceptive use. In their study, they

21

also found that an educational intervention based on HBM produced statistically
significant increases in condom use among adolescent\.
Jn 1988. Rosenstock, Strecher and Becker presented the argument that the

Health Belief Model and Social Cognitive Theory were connected. They reported that
both involved the concepts of threat. expectation of outcomes and incentive or
motivation. They also proposed that. while HBM did not specifically deal with self
efficacy, this concept was implied under perception of barriers and the need to weigh
barriers with the outcomes of preventive behaviour.
Social Cognitive Theory (SCT) was introduced hy Albert Bandura. Social
Cognitive theorists believed that
.... people are neither driven hy inner forces nor automatically shaped
and controlled hy external stimuli. Rather, human functioning is

explained in terms of a model of triadic reciprocaJity jn which
behaviour, cognitive and other personal factors, and environmental
events aJI operate as interacting determinant~ of each other (Bandura, 1986, p.
18).

People were seen as having an active role in the determination of their hehaviour~,
thoughts, emotions while the impact of the larger system was also acknowledged.
According to SCT (Bandura, 1986) human nature is defined by the foJlowing
human capabilities: the abilities to (]) use and interpret signs and symbols, (2) use

22
forethought. (3) learn through vicarious experience. (4) regulate hehaviour through
personal standards and self evaluation. (5) he self reflective - to reflect on cognitions,
act on them either physically or in an anticipatory way, explore results and change
behaviour accordingly. Four processes of behaviour change arise from these five
capabilities. People can change their thought~ and behaviours through direct

experience and mastery. through modelling or indirect experience. through social
persuasion, ie. social judgements, and through inference based on interaction with
self, others and environment. In the fourth case belief in the physical ahility to behave
in a certain way plays a major role. These four experiences are also the ways to build
self-efficacy (Bandura. I 989).
The belief in self-influence appeared to he a cornerstone of Bandura 's theory.
A major aspect of self-influence/self-determination is perceived self-efficacy.
"Perceived self-efficacy is a judgement of one's capability to accomplish a certain
level of performance ... " (Bandura, 1986, p. 391). There are three major ways in
which self-efficacy impacts our functioning. First, it has an impact on the choices we
make to hehave. Second, is the impact on motivation. Expectations of ability to
perform Jeads to higher level of effort. Third, self-efficacy impacts a person's

patterns of thought. Bandura asserted that when problem-solving, a person with a high
sense of efficacy can focus all attention on resolution.

21

In an interview with Evans ( 1989), Bandura spoke to the relevance of
perceived self efficacy to health promotion. When questioned ahout this relationship,
Bandura asserted that efficacy had an impact on whether a person would consider
and/or adopt certain health preventive hehaviours. Efficacy could affect the feeling of
benefit from treatments as well as maintenance of preventive behaviours.
Social Cognitive Theory and, more \pecifically, the concept of self efficacy
has been used in many research efforts around change in health hehaviour, including
HIV preventive hehaviour<;. Studies have shown the impact of perceived self-efficacy

on health behaviour, for example, Beck & Lund (1981) and O'Leary et al (1992).
Bandura, Adams and Beyer ( 1977) showed that interventions promoting positive
change in perceived self-efficacy were effective in helping people change behaviour.
In this case the behaviour was related to a phobia of snakes. They showed that those
who attended the interventions designed to model and/or practice assertive behaviour
with snakes had higher perceptions of personal efficacy than the control group.
In a study on patients living with Chronic Obstructive Pulmonary Disease,
Kaplan, Atkins and Reinsch ( 1984) discovered that specific efficacy in the form of

exercise instructions led to increase in exercise behaviour. Changes occurred due to
increased feelings of efficacy for exercise brought on by the instructions about the
exercise. According to the researchers, this study supported the role of self-efficacy
and SC'T generally.

24
Jacohs, Prentice-Dunn and Reinsch ( 1984) found that eflicacy, outcome
expectancies and ,elf-awareness all had a part to play in persistence. Their study
revolved around pcr,1\tencc after a task was failed. The strongest interaction
occurred among those who had hcen exposed to .self-efficacy manipulations as an
interventive strategy after the preliminary failure. These outcomes lcc.J the authors to
state that ". .. self-efficacy expectancies were the best predictor or persistence. "(p.
333). Acknowledging the impact of outcome expectancies on perception of efficacy
1s important. The perception of negative consequences to using condoms was related
to a decrea\c in ju<lgernent of ability and less practice of preventive hchaviour.
Therefore, the expectations of the negative aspects of using condoms, such as the lack
of stimulation, led to a dccrea\e m the perception of ability to consistently use
condoms. Exploration and identification of the componentli of self efficacy with
respect to safer sexual behaviour wa, necessary (Abraham and Sheeran, I 994).
Some research studies were observed to combine related components of
several theories as their framework. One such theoretical model was the Al DS Risk
Reduction Model (AARM). Elements of such theories as HBM, TRA and SCl were
integrated to better explain AJ OS preventive actions (Mantell ct al., 1997).
According to Mantell et al., the important factors of ARRM were knowledge of risk
and methods to increase satisfaction of low risk behaviours, perception of risk,
weighing of perceived costs and henefits of low risk act4i, hclief in ahility to perform

2~
the behaviours, emotional states. an<l social factors \UCh as group norms.
Accordingly, behaviour change occurred when (a) the behaviour was identified as
high risk~ (b) a commitment was made to change the hehaviour~ (c) the intent to
behave wa~ transformed into action . The movement from one step to another was
mediated hy achieving the goals of the previous stage and belief in the ahihty to
achieve the goals of the next stage.
Factors and Barriers Affectin~ HIV Preventive Attitudes and Behaviours A
major area of study seen in the literature was exploratory in nature Researchers
attempted to determine HlV knowledge, attitudes, behaviour and contributing factors
and barriers to prevention in populations Research focu5ed on exploration of the
psychosocial factors influencing behaviour (Kowalewski et al., 1997). This focus
seemed to come from an attempt to define all factors affecting preventive behaviour so
that these factors could be used or worked on within development of interventions.
Measures of HIV knowledge were taken in populations to explore levels of knowledge,
myths in knowledge and the impact on knowledge on undertaking preventive behaviour.
Attitudes about HIV and ideas around H1 V prevention were measured to understand
how attitudes affected behaviours and how they could be used in the promotion of
preventive behaviour Researchers tested for factors that they hypothesized were
important in J UV prevention
When adolescents were first identified as at risk for HIV infection, an immense
Canada Youth and AIDS Study was designed to explore knowledge, attitudes and

26

behaviour in Canadian youth nationwide King et al. ( 1989) reported that more than
38,000 youth from across Canada were surveyed Six surveys examined gender, age,
knowledge of A l DS, and behaviour considered to pose risk The primary conclusion of
the study was that AIDS was a threat to youth There were eleven conclusions as
follows
1. There was a threat to youth for HIV contraction

2 Responsible behaviours were essential in prevention.
3. Accurate knowledge was imperative to behaviour change
4. All people have a right to accurate information and it was the responsibility
of the community to educate in the area of AfDS .
5. Links should be made between AIDS and STD's in educational programs.
6. Dissemination of information should be done by those knowledgeable who
use appropriate language for this population.
7. A number of adolescents were sexually active and educational approaches
should reflect this.
8 Evaluation of educational interventions before they are implemented was
essential
9 Sub groups of this population at highest risk were identified as needing
priority status in preventive strategies.
I0. AIDS education was considered most effective when it was in a larger
program that addressed several concerns at the same time.
I I. Educational programs needed to encompass accurate presentation of

27
information, development of skills necessary to carry out preventive
behaviours, promotion or tolerance of and compassion for persons living
with I llV and AID~ (pp 14 3-144 ).

Thurman and Franklin ( 1990) studied the knowledge of Al DS and perception of
health threat from AIDS in college students at lhc University or fassachusetts/ Amherst
The~ stated that students were well informed about AIDS and preventive behaviour
They also cxpre~s<..d that personalizing the threat of I IIV was necessary before behaviour
would change. The discordance bet\,een knowledge and behaviour supported the case
that merely educating people about AIDS and HIV was not enough to lead to behaviour
change. One suggestion. made by the researchers was that "future research , on and off
college campuses 1s needed to determine better the perceived susceptibility to the
disease of students, and the general public'' (p. 183 ). ·1he role of perceived
susceptib1hty was a major concern in primary prevention and is explored further in the
next section of the review. While Thurman and Franklin showed students to be fairly
knowledgable about AIDS, Bro\\ n, DiClemente and Beausoleil ( 1992) stated that many
studies have shown significant gaps in kno\\ ledge for this group.
Knowledge has been tested in many situations for a variety of populations and
groups, including homosexual men, intravenous drug users, college student". \ outh and
the heterosexual population. Comparative studies have also been done between groups.
For example, Brown~DiClemtc & Beausoleil ( 1992) compared the knowledge, attitudes
and intentions to practice safer sex between young adolescents who v. ere sexually acti, e
and those who wcren 't These researchers found significant differences between the'" o

28
groups. For example. measured difference was more pronounced in young boys who
were sexually active versus those who wer c not. Significant differences bet ween males
and females also occurred. The authors hvpothesizcd that young people who were
sexually acttve felt a need to minimize their risk and so v.crc not really integrating the
information provided They postulated that prevention strategies in the schools were
focused on abstinence which led to the possibility of alienation of the sexually active
group. The researcher~ stressed that cflorts must be continued in promotion of positive
attitudes. knov. ledge and beha\ iour of young people who were not engaging in sexual
intercourse. while efforts to aid young people who were sexually active needed lo be
·stepped up' .
Exploratorv stud1e around the topic of behaviour focused on establishing risk
behaviours for populations, identifying barriers to preventive behaviour and, more
specifically. condom use. Butcher, Manning and ()'~eal ( 1991) identified risky
behaviours in college students. These behaviours included alcohol consumption (people
reported sexual intercourse when intoxicated) and sexual intercourse with multiple
partners, that seemed to be defined as serial monogamy. The students reported having
sex with only one person at a time. but having an average number of three partners in a
year. Most of the students said they did not use condoms in these interaction~. In their
summary of the Canada Youth and AIDS Study, King et al. ( 1989) defined more risky
behaviours. Their findings showed that most older youths had engaged m sexual
intercourse and usually with more than one partner. It was also rcportC' I that much of
this interaction occurred without the use of a condom and spermicide. The study al. o

29
showed a that youths were engaging in anal intercourse
Krull ( 1994) defined different factors that affected risk behaviour. Krull
perfonncd two studies on data gathered through a national sun cy in the United States
In study one, she attempted to assess the interaction of levels of education. liberal sexual
attitude and sexual promiscuity and. in study two, she related these factors to risk for
AID

Krull found that levels of education led to more liberal sexual attitudes which

were associated with sexual promiscuity in the group 'A ith whom she was dealing This
pointed to the increa~cd risk or AIDS brought on by sexual promiscuity Therefore.
people with higher levels of education \\-ere presented as having more liberal sexual
attitudes. These more liberal attitud\!s, coincided with increased promiscuity. And,
finally, increased promiscuity indicated a higher risk for contraction. Krull's study
supponed the case that higher levels of education had not protected people from /\IDS,
and, as a matter of fact, may have increased behaviours that were related to contraction.
The report also supported the case that all groups of people should be addressed when it
comes to prevention development
The findings in the abov(, studies point to several behaviours that needed to be
considered when designing preventive strategies, particularly with older adolescents and
young adults. Those behaviours included sex with multiple panncrs, lack of condom
use, sexual intercourse under the influence of alcohol, and anal intercourse. Other
behaviours of interest included lack of discussion of preventive behaviours with partner
and lack of provision of and/or suggestion to use condoms ( Caron, Da\ is, Halteman. &
Stickle, l 993~ O'Leary el al., 1992).

10
Condom use 1s a behaviour that was a common toptc rn the literature
Researchers looked at such things as predictors of condom use, use or condoms in
partnerships and changes in condom use Batli ct al ( 1989) looked at factors that

-

influenced the use of condoms in male heterosexual college students. N

JOS . rn this

study, the subjects were given questionnaires measuring such factors as numbers of
partners, intent to use condoms. and attitudes lO\\ ard condom

The authors found that,

in the two months prior to the study, 67o/o had at least one sexual partnc1 , 14% had two
sexual partners, 5% had engaged in intercourse with three partners and 3 33% with more
than three. These subJcc..ts repo11ed that they did not consistently use condoms and that
their use of condoms was mostly for the purposes of contraception. Many of the
subjects did report a v. illingness to use condoms, but expressed such barriers as level of
comfort in buying condoms and lack of communication between themselves and their
partners about preventive behaviours. Therefore, factors that impacted the use of
condoms were the provision of another fom1 of contraception, lack of communication
with partner, and social level of comfort with condoms ( which could point to an
underlying level of comfort in sexuality generally).
Other useful information in the area of condom behaviour involved change in
condom use. Catania. Stone, Binson, & Dolcini ( 1995) reported that when condoms
were used, they were not used consistently. They looked at heterosexuals from several
groups including high risk groups. The researchers reported that lack of consistency had
implications for I IIV contraction given that safer sex involved consistent use of
condoms.

The authors also reported that condom use among heterosexuals had

JI

increased from 1989 to 1992 and then had begun lo le\el ofI According to the
researchers, the results of the 5ludy pointed to a need to seek new avenues to promote
condom u 'C
Modification and Promo.tio.n

or HIV Pr y_CntILc..Altitu.dc_s_and Behaviours Jn an

exploration of literature, this researcher discovered several studies that provided valuable
information for development of int en cntions These included de\ clopmcnt int en cntions
based on theory driven incentives to promote condom use (Grube, 1ayton & BallRokeach, 1994: O'Leary ct al., 1992), interventions that integrated education and
motivation for positive attitude and behaviour change (Fisher et al , 1996, Moskal,
1991 ) : preventive strategies that addressed communication issues ( DeBro, Campbell &

Peplau, 1994), and interventions based on small group interaction (Tudivcr et al , 1992 ).
One set of interventions, developed by Tudiver et al ( 1992), was based on small
group education sessions for gay and bisexual men. Three groups were involved in the
study. T~o intervention groups in which subjects were either involved with a single
educative session guided by a peer volunteers, or a \enes of four sessions led by trained
counsellors. This study supported the case that single ses\1on interventions \verc not
necessarily less effective that multiple sessions and that peer helpers could be useful in
carrying out interventions.
Fisher et al. ( 1996) designed an intervention using the lnformation-MouvationBehaviour Skills (1MB) model of behaviour change. In the 1MB "information and
motivation are thought to affect the use of risk reduction behavioural skills that arc
necessary for initiating and maintaining patterns of AIDS risk beha\ iour change" (p

32
11 '5) Subjects in an intervention group were presented with deficits in information

around AIDS pre\ entivc behaviour information They attended small group discussions
led by peer leaders in \\ hich the peer leaders acknowledged the same ncgat ive attitudes
and beliefs as presented in the information session. In these scs~ion the subjects
problem-solved around ways to deal with the underlying factors of the negative beliefs.
Later subjects attended a large group session dc~igned to reinforce any changes made
due to the interactions in the small group and lo present information around changing
behaviour and maintaining it in their social circles through watching a video, modelling,
and skills building. Subjects attended small and large group discussions aller this in
which they were taught skilJs that could be used in consistently displaying preventive
behaviours. The subjects in the intervention group were shown to have changed
behaviours and maintained those changes over time.
Moskal (1991) implemented an AIDS curriculum over a tv.o month period with
college students who were predominantly First Nations This study was also pre/post
intervention/control design. Subjects took part in such activities as attending lectures,
performing opinion polls and discussmg controversial issues in the area of HIV
prevention. The course included a variety of methods and activities ~uc.h as lectures
involving relevant videotapes, anonymous opinion polls, noajudgemental discussions of
controversial issues, and a classroom demonstration of the use of the condom. Role
playing was used as well as development of posters or a video tape. Results showed that
the curriculum was eflcctive in greater change in knowledge. attitudes toward lo\, risk
behaviours and towa1 ds those who are J II V pos1t1ve

33
The three studies d1scu<i~td in thi~ "iect1on showed valuable strategics in
prevention. In all three studies. researchers were able to report a level of cflicacy in
promoting positi\ e behaviour and/or attitude change Common processes in the above
studie') mcludcd : the provision of information that was supported by strategies to
promote change, interactive components in which the subjects \\ere involved in
actt'vitics: and. creative activities that could increase interest and involvement in the
interventionc:; All studies targeted condom u c as a preventive bchaviow, hO\\ ever, not

many studies integrated the provision of abstinence as a realistic choice
Challenges and Solutions in HIV Primary Pre\ention

Stevens-Smith and Remley (1994) supplied some important considerations for
SCT in prevention planning. The first was that correct information ahout HIV must
be given. This included pattern~ of transm1s\1on and explanation of HIV and AIDS.
The second consideration stressed the importance of helping people develop soc..1al and
regulatory skills. Here they asserted that it was important to address people's
responsibility in controlling their attitudes and hehaviours. A third consideration
offered continued aid in further development of skills and encouragement in
contmuance of personal control. The final consideration was the necessity of social
support to improve chances that preventive behav10ur is maintained. These factors
reflected some of the challenges to HIV prevention. The next section of this litcratur~
review will present information regarding various challenges to HIV prevention and
suggested strategies to deal more effectively with the challenges.

34

Sexual Behaviour HIV transmission is directly related to sexual bchaviou1.
although there arc other modes of contraction Sexual behaviour is a complex process
that integrate~ cultural beliefs, interpersonal power issues, gender role prescriptions, and
personal responses such as fear or reJcction and denial (Evans, I989, Wilson & Stewin,
199:1) The contraction of I IIV does not occur in the insular world oft he individual.

Due to the way I IIV is transmitted, the social role in contraction is as important as the
personal characteristic\ of the individual If I II\" is transmitted through sexual activity,
then the interaction involves at least two people There is more to consider than the
beliefs and ideals of the indi\~dual In the development of interventions it is essential to
design interventions geared toward both of the partners, in two couple relationships
This can impact the norms established within the dyadic relationship ( DeBro ct al.,
1994: Fisher et al., 1995).

Trussler ( 1994) pointed to a consideration that must come into play in the
interaction of heterosexual couples. This factor related to sexual negotiation and the fact
that the means for a woman to protect hersel'' from IHY is usually controlled by the
male, ie. condom use. This was considered a power differential with which women have
to deal in relationships Caron et al. ( 1993) also reported that women with traditional
attitudes toward their gender role were more likely to forego safer sex. Women have
been found to be less comfortable with sexual issues than men and to have felt less
responsible for thetr sexuality and less powerfol in a sexual encounter (81yan et al. ,
1997). Many of the researchers presented in this section attested to the fact that gender

differences were a part of sexual behaviour and communication and that inter\ entions

needed to address this factor
The larger system in which a couple functions also impacts sexual interaction
Perception of opinions of others can have an influence on the motivation to indulge m
high risk activit1es (Baker et al., 1996)

tv1any studies and programs professed that it

was essential to teach communication skills and to help promote assertive behaviour in
the sexual practices of target groups (Bl)·an. Aiken & West. I 996~ I 997). This included
such activities as role plays in which a person interacted with a proposed sexual partner,
practice in facing social group pressures around <;exual intercourse, amJ practice of

skills necessary for condom use (Fisher et al., 1996). Cotton, Higgins, Person, &

Darrow ( 1994) descnbed how the powerful influence of peers was used . They spoke of
role mode) stories in which members of the communhy relayed stones of their own
struggles and successes with safer sex. prevention and HIV. These stories were printed
in brochures and flyers and distnbuted by other members in the group. The fonn of
distribution allowed for the inclusion of an interpersonal dimension of interaction. The
credibility of a person who was experiencing like situations as those they were
attempting to reach could help those people make positive changes in attitudes and/or
behaviours (Lief, 1989).
Sexual behaviour contains an individual component that 1s comprised of emotions
and experiences around sexuality Abraham and Sheeran ( J 994) defined sexual
behaviour as very different from other health behaviours because of the high emotional
and arousal content, among other things. They proposed that individual differences in
emotional responses to sexual stimuli were the real predictors of safer sexual behaviour.

36

They expressed that any interventions dealing with sexual behaviour must address the
underlying emotions and the individuals, methods of dealing with emotion.

Cultural/Group Considerations Each group or culture is unique and has its own
patterns of behaviour, jargon. boundaries, ideas of acceptability, and rules for interaction
(Baron & Byrne, 1987). It would seem reasonable to assume that this uniqueness must
be considered in any type of preventive strategics that are designed .
One example of such uniqueness and the need for consideration of that
uniqueness in designing preventions is language. Terminology used by those who
disperse infom1ation for consumption by groups is not necessarily known or understood
by those groups (Canadian Public Health Association, 1993). This must be taught and
learned AJong with this comes the jargon which is used by various groups. If a person
is to deal effectively with a particular group, it is immensely useful to use the jargon and
patterns of speech of that group .

An essential aspect of addressing the target group is to use group appropriate
language Language defines who we are as a people and as individuals To use jargon
with which people are unable to associate is to isolate them from the very beginning
(Lief, 1989). It would seem unlikely that any intervention that used inappropriate
language would prove to be effective One way of getting to the language and
manipulation of language of a particular group is to involve the group as ·experts' in the
understanding of the dynamics of their respective groups. An integral pa11 of those
dynamics is the language.

37

One important strategy for dealing with the difficulties posed by group 01 cultural
barriers is to include group members in the planning. implementation and evaluation of
the programs Lo be used with their group (Cotton et al., l 994, Orians et al.. 1995,
Sand fort & Zessen, 1992) The abiltty to serve the needs of the populalion can be
referred to as cultural competence. The CDC. 1996, provided an in-depth definition of
cultural competence
Participation of client populations throughout the process of designing and
implementing programs helps as':)ure that the program will be acceptable to
the persons for whom it is intended. For the purposes of this document,
cultural competence is defined as the capacity and s"1II to function effectively
in environments that are culturally diverse and that are composed of distinct
elements and qualities . The fact that many organizations in a community are
directly controlled by the members of the community will lead the organization
to reflect the beliefs and interactions of the community as well as some oft he
difficulties which must be addressed by anyone targeting that community for
interventions (p. I).
Bryant et al (] 992) suggested that the involvement of members of the target group can
increase the possibility of cultural competence This group presented the target
populations as the consumer and thus "the inclusion of the consumer allowed
participants to provide input about the severity of the problem, to discuss the
environmental and social conditions that affect the problem, and to take an active role in
desired change" (p. 28) Therefore including those for whom the interventions are being

designed can increase the role consumers play in their own change.
Denial and Underestimation of Risk The research and resources on denial,

perceived vulnerability and perceived nsk, showed that these three concepts were
directly related to each other Several researchers discussed the importance of
addressing these three concepts in prevention planning (Wcmstein, 1989, Cohen &
Bruce. 1997~Fisher et al., 1996~O' Leary et al., 1992).

Sandfort and Zessen ( 1992) reported that denial downplays the perception of risk
to oneself, one, s system and the larger system (societ, ). This lessening of perceived risk
can lead to less tendency to practice those behaviours that are essential in living a
preventive lifestyle These researchers gave several consequences of denial of risk .
First, people may block out prevention messages and other valid information. The
ignorance brought on by this exclusion could lead to increased risk. Second, while
people may be unaware of necessary messages and information, they may not translate
this awareness into positive behaviours. Sandfbrt and .lessen discovered that subjects
who heard the messages but tended to downplay them were less likely to take on future
preventive behaviours. People not only downplayed the information they were given but
also downplayed their risk.
People who have been at risk and who reason this risk away don't
acknowledge the fact that they have been at risk. These people are
often less worried about the possibility of bemg infected In general,
the stronger the tendency to reason away the risk of infection, the less
often subjects worried about their future risk of getting infected (p. 69).

19
This type of denial has ramifications for HIV prevention because all of the infonnation
that can be provided will not be enough without helping people overcome the underlying
tendency to deny/reason away the nsks being relayed to them.
Sandiort and Zessen ( 1992) found that the changes in sexual behaviour were
"related to whether one knows people who are I JIV infected or who have AIDS. As a
consequence of personal confrontations v.1th HIV/AIDS related issues, the awareness of
potential risk might be heightened" (p. 71 ). These researchers stated that interventions
should include an element designed to decrease denial of risk. They proposed that an
effective element would be to make v1s1ble to target groups that I IIV affects people who
don't belong to traditional risk groups Orians, Liebow and Branch ( 1995) also
suggested that interaction with those who are I HY positive or full-blown AfDS can
make the threat of contraction much more real The Canadian Public Health Association
took a clear stand on inclusion of infected people in prevention. They proposed that
"programs and services should be developed and delivered in consultation and/or
partnership with people aftected by/ infected with AIDS" (CPHA~ 1993, p. 2).
Sandtort and Zessen did caution, however, that for some people, exposure to those who
have HIV or AIDS can be so traumatic as to increase the potential for denial. In this
case, other interventions were considered more appropriate.
In order for individuaJs to make a realistic estimation of their risk several things
must occur. They must have the most up-to-date information and not be dealing with
myths and partial truths (Cohen & Bruce, 1997). The risk must be personalized in order
for them to apply it themselves (King et al, 1989, Thunnan & Franklin, 1990)

40
f ndividuals must have a realistic perception of their \'Ulnerability to the risk .

"Perception susceptibility refers to the likelihood of experiencing personal harm if no
action 1s taken" (Weinstein, l 989) People, particularly adolescents and young adults,
seem to have difficulty determining the11 susceptibility to J IIV contraction. Seppa
( 1997) spoke to this issue when he addressed the concept of invulnerability. This
perception of invulnerability can come from denial. misinformation, confosion, and the
general perception of invulnerability to insult expenenced by youth .
Cohen and Bruce ( 1997) argued that people were not unrealistic in their perceptions of
their vulnerability to contraction. They suggested that in comparing perception of risk to
epidemiological forecasts of risk in certain populations, subjects were actually
overestimating their potential to contract HIV. They suggested that the risk for
contraction in the heterosexual, non-IOU population was negligible. They did, however,
report that statistics for mortality due to AIDS was very high m certain young adults,
and that it was very clearly a major health risk in particular groups.
Research Hypotheses
The extensive review of the literature led to the structuring of the research
hypotheses Two hypotheses were defined for the study represented in this the~is
These hypotheses revolved around assumptions of efficacy in two types of interventions:
an affective intervention in which the subjects attended a presentation given by a person
living with HJV, and a second tntervention in which groups of subjects designed
interventions for their peer group. The hypotheses were expressed as follows :
1. An intervention, hereafter referred to as Intervention One, involving presentation of

41

accurate information about HIV and AIDS. interaction with a young adult who is HIV
positive, and developing goals around preventive behaviours will be effective in
promotion of positive change in attitudes and behaviours around HIV
2. An intervention, hereafier referred to as Intervention 2, in"olving presentation of
accurate informatton about HIV and AIDS, involvement in planning a prevention
program, and developing goals around preventive behaviours will be cfrective in
promoting positive change in attitudes and behaviours around Hl V.
Statistical hypothei;;es for attitudes and behaviours can be found below:

Ho: µai = µ31 where i, j, = 1, 2 or 3 and i

I

j

H1 µm ;1; µ:i.1 where i, j, = I, 2 or 3 and i * j
In the above hypotheses, the null hypothesis (I 10) would be that the means of all three
groups are equal To reject that hypothesis (H 1) would lead to the conclusion that the
means of groups are significantly different. In the notation :i represents attitude.

Ho: µhi = µhj where i, j, = I, 2 or 3 and i

-:1:

j

H1: µhi -:1: µhi where i. j, = I. 2 or 3 and i I j
In the above, means are all equal if the null hypothesis (H0 ) is retained. The means are
considered to be significantly different if the null hypothesis 1s rejected. Behaviour is
represented by the figure h· In both sets of hypotheses, rejection of the null hypotheses
would mean that significant change had occutTed in attitudes and behaviours between the
pre and post tests

42

Summary
This chapter contained a presentation of information resulting from an
investigation of the literature around HIV and prevention An historical overview of
HIV and relevant ep1demiological information was provided. A presentation of the types

of research being done in the area of HIV prevention was given and common challenges
in the literature were explored . The final section in the chapter presented the research
hypotheses of the study, that interventions involving interaction with a person living with
HIV or involvement in planning a prevention program would be effective in promoting
positive attitudes and behaviours. In the next chapter methodological issues and
procedures used in the study are explored

43

CHAPTER 3

METHODOLOGY

In this chapter. recruitment. participants involved in the study. the process of the
study and types of statistical analysis that will be done arc described. The settings
involved, all materials necessary for the execution of the interventions, the procedures
followed in the administration of tests and the approaches used in the interventions are
presented The demographic. attitude and behaviour questionnaires arc described. This
description includes types of items used in the questionnaire package, examples of items
in the attitude and behaviour scales, and references to those inventories and
questionnaires in the literature used as models by the researcher in item preparation. An
overview of the procedures to be used in the data analysis is also presented

Subjects
Subjects for the study were canvassed, through convenience sampling and a
snowball like effect (Palys. 1992). from the student population at the University of
Northern British Columbia (UNBC), in Pnnce George, British Columbia Due to the
small numbers of students at the institution (n- 2500). and to low participation at other
educational events, the researcher decided to collect subjects through personal contact.
Students were contacted by visiting a first and third year psychology class. as well as the
majority of apartments in residence v.,1th the aid of the Resident Assistants. The
investigator described the type of research being done, and left sheets for the students to
fill out and return, anonymously, to Student Success Centres The sheets contained

44

information about appropriate times for the pretest and space for subJecls to enter their
name and contact number An option was given to contact the researcher by a private
phone number if the subjects were not comfortable with the idea of passing their sheets
to the receptionist at Student Success Centres. The total number of participants for the
study were 3 I (N · J I ).
Setting
The in\ estigator held the interventions on the prcm1~es of UNBC and used a
large lecture theatre for the administration of pre and post tests The decision to use
such a large hall was an attempt to increase confidentiality and fi.irther demonstrate to the
subjects that confidentiality would be a top priority. In a larger hall there was more
room for students to spread themselves around and the researcher instn1cted students
about the space necessary between them. A lecture hall was selected for Intervention I.
Intervention 2 necessitated a room v.'ith movable chairs These rooms were empty of all
materials with the exception of those brought m by the researcher.
Research Instruments
A questionnaire package was used to measure attitudes and behaviours of the
subjects pre and post test . The package consisted of a questionnaire measuring
demographic information (Appendix B), an attitude questionnaire (Appendix C) and a
behaviour questionnaire (Appendix D). The researcher designed the questionnaires to
specifically reflect those attitudes and behaviours to be measured. Piloting of the
questionnaires occurred in a first year and a fourth year psychology class. The
investigator also gathered feedback from coJleagues in a master's level research seminar

45

course. As suggested by Sax ( 1989), the feedback concentrated around choice of the
most appropriate questions, analysis and critiques of the questions as thcv had been
presented, and identification of any areas that may have inadvertently been omitted.
Demographic Questionnaire. The demographic questionnaire contained openended questions related to demographic information necessary for interpretation of the
results of the experiment. The measure was used only in the pretest. Types of
information sought with this questionnaire included the age of subjects, gender, sexual
orientation, measure of sexual experience, and degree and type of I ITV knowledge. The
questionnaire also measured the amount and type of communication about HIV between
the subjects and family, friends and sexual partners. These considerations added to the
understanding of the sample used in this study They also provided measures on some of
the factors that contributed to HIV preventive attitudes and behaviours in the literature,
such as gender as reflected by Caron et al. ( 1993 ): DeBro et al., ( 1994 ): and Catania et
al. ( 1995)

Attitude Questionnaire. The second test was composed of questions designed to
measure attitudes toward HIV prevention. The attitudes to be measured were selected
from such sources as Thunnan & Franklin ( 1990): O 'Leary et al. ( 1992): Torabi &
Yarber (1992). Caron et al (l993t and Dancy (1994). The attitudes to be addressed
were reflected in subjects' responses to the following
I.

HIV infection is limited to homosexuals, drug users and those who receive
blood products.

2

Condoms are not a necessary aspect of sexual intercourse.

46
3.

Knowing your partner is the best prevention of HIV contraction

4.

Abstinence is never practised as a preventive measure against
the contraction of HIV

5

Co1nmunication of the need to practice safer sex leads to
reluctance or r~jection.

6

A person is powerless to protect him/herself against HIV
contraction.

The first section of the attitude questionnaire contained a selection of open-ended
questions which were provided to measure attitudes that may not have been indicated in
the Likert scale format Mantell et al. ( 1997) suggested that open-ended questions
allowed subjects to respond without the confines of a Likert type scale and without being
led by a series of responses in which they must fit. These items were placed at the
beginning of the questionnaire to lessen the 1mpact of the more stn1ctured questions on
student responses (Sax, l 989)
An agreement attitude scale was used in the second section of this questionnaire.
Responses were made through a 6-point Likert scale range'' very strongly agree",
"strongly agree", "agree". "strongly disagree'', and "very strongly disagree" . The
researcher const1ucted Likert scale questions from a list of attitudes drawn from the
literature and formulated to reflect change if the interventions were effective.
Examples of attitude items in this questionnaire included ·
1.

Using a condom during sex can be erotic

VSA SA A D SD VSD

47

2

No one practices abstinence as a prevention
for HIV infection.

VSA SA A D SD YSD

Response options of ·'not applicable" or "undecided" were omitted to allow various
levels of agreement to be measured.
Behaviour Questionnaire. The third questionnaire measured specified verbal
behaviours relevant to safer versus unsafe sexual practices These behaviours were
gathered from such sources as Butcher et al. ( 1991 ); Caron et al. ( 1993); Dancy
( l 994 ): and van der Plight and Richard ( 1994). The placement of questions and
procedure for the construction of items was the same as in Questionnaire 2. In this scale
the measurements range was "every time", "sometimes", "once or twice", "never", and
"not applicable" and they were a1Tanged in a 5-point Likert Scale. Examples of
behaviour items in this questionnaire included
IN THE PAST SIX MONTHS I HAVE ....
l.

depended on my partner to

ET ST TO N NA

provide protection.
2.

asked a partner if they had been
tested for HIV

ET ST TO N NA

The investigator used this type of measure instead of a "yes/no/not appJicable'' measure
in an attempt to gauge possible nuances of behaviour. rt was believed that an uncommon
occurrence might lead to an answer of ''no'' or "not applicable" thus disallowing a proper
measure of that behaviour.

48

Procedure
One hundred students \Olunteered to be part of the research and sixty or these
completed the pretest Thirty-one students completed the full study. Reasons for the
lack of extended involvement may have included such things as student schedules,
decision not to take part after complettng the questionnaires, or Jack of interest The
subjects were randomly assigned to one of three groups. the Control group, N - 9:
Intervention 1, 1\J -; IO: Intervention 2, N = 12. Before administration or the pre tests, a
subject number and a group letter were printed on each manila envelope The envelopes
were placed on a table m random order and subjects were asked to pick an envelope as
they came in. Therefore students v.,ere randomly assigned to each of the three groups
The subjects completed a consent form before they began the questionnaires On the
back of the consent form, they were asked to choose appropriate days out of those
provided by the researcher for the interventions. The need to have the interaction of as
many subjects as possible, negated a plan to have equal representation by gender.
Prior to opening the envelopes the investigator gave basic instructions. The same
instructions were placed on overhead for subjects to read as they were going through the
test One of the instructions involved the subjects placing a unique character on the
outside of the envelope which only they could identify. This ensured that subjects could
access their envelope if they forgot their subject number and group letter. Students'
names were not entered anywhere on the questionnaires or the envelope. The researcher
further instructed the subjects that the questionnaire was to be completed and replaced in
the envelopes in the order given When subjects finished , they returned the envelope

49
with contents enclosed to the investigator or assistant and were reminded that they
would be contacted when the interventions were to take place. In an effoJ1 to
accommodate as many subjects as possible, the pretest occurred three times so that
subjects could choose a time slot that swted their individual schedules
All subjects in the Control group were contacted and informed that they were
members of the control group. The researcher explained their responsibility as members
of the Control: ( 1) They were informed that they need not attend any intervention. (2)
They were expected to return for the post test 111 early April. Due to the fact that many
of the subjects fed back knowledge of control groups and their purpose, the researcher
explained to all the reason for the use of control groups. This explanation provided
general terms with no reference to the study in which the subjects were taking part.
Intervention I necessitated contracting with a person who Vvas living with HIV.
A young heterosexual woman who had contracted the virus during her second pregnancy

agreed to speak to the subjects involved with Intervention 1. The researcher accessed
the volunteer through the AIDS Society in Prince George. This young woman gave an
account of her personal journey through the process of discovering that she was HIV
positive, and how she was dealing with the situation. She also discussed more general
aspects of HIV prevention which would be comparable to the information provided for
the Intervention 2 group. The researcher met with the woman several times before the
intervention took place to discuss the context of her presentation. The intervention
lasted two and a half hours and was set up in the following way
1.

Presentation of the facts about HIV:

so
A. Definitions of JIJV and AfDS
B

I lo~ HlV is contracted.

C. Ways Lo prevent contraction.
D. Cultural barriers to preventive strategics.
E. Gender differences in the area of HlV prevention.

F. Importance of peers in decision-making.
G The role of self-awareness in living positive lifestyles.

H. The importance of keeping up with current information and the
usefulness of setting personal goals
2.

The speaker presented the process of contracting HIV from her
perspective She addressed her own feelings and diflicultics around being
HIV positive and some of the lessons that this has taught her.

3.

Time was provided for questions to the speaker, the representative or the
researcher.

4.

The subjects were asked to set reasonable goals for themselves
based on the information with which they were provided and their
own ideas of preYention.

Instruments needed for the second intervention consisted or infonnation gathered
by the researcher from various organizations. The majority of the resources were
obtained from the National AIDS Clearinghouse. a branch of the Canadian Public Health
Association concerned with dissemination ofinfonnation on HIV and AIDS . The
resources used included general infon11ation about the l llV and AIDS. causes_

51

symptoms, prevention strategies (mostly focused on safer sex), and blocks to preventive
behaviours These were the only instruments necessary for the second intervention with
the exception of an overhead projector. paper and pens. The subjects in this intervention
participated in planning their own prevention. They were divided into three groups and
asked to consider what they felt would be helpful for university students of their age
group, 18 - 25 ) rs . based on the verbal and visual information provided for them . The
process progressed as follows:
1.

Presentation of the facts

n,is was done the same way and using the same

materials and information as the first intervention.
2

Outlining of strategies based on the following procedure, and with
access to information sources which would help them make their
decisions:
A.

Definition of the needs of young adults at the university
level

B

An outline of strategies which they felt would be most
successful in helping members of this group meet their
needs.

C.

Projected outcomes of the intervention

D

Rationale for the use of these strategies

3.

Presentation of the intervcntion(s) to the other groups involved

4.

Discussion and questions regarding any of the information or part of
the process

52
5.

Subjects were instnJcted to set reasonable goals for themselves to help
prevent contraction.

At the end of both Intervention I and 2. a brief discussion occurred as a point of
debriefing Subjects were also reassured that if they had any questions or wanted to
speak more about their experiences with the intervention they could feel free to contact
the researcher
The post test occurred seven v.ceks after the pre test had been administered, and
two weeks after the interventions occurred. The researcher used the same attitude and
behaviour questionnaires in the post test as had been administered in the pre test. The
order was reversed to further decrease the potential of students responding from memory
of the pre test. An open-ended question asking subjects to report any perceived impact
of the interventions on them was mcJuded in the post test and the demographic
questionnaire was not included The investigator informed members of the control
group that it was not necessary to respond to the added quesuon. Some members of the
control responded an}".vay. The researcher' s assistant collected the packages and they
were locked in a secure filing cabmet in a locked office until the data could be entered
onto a database
For both interventions there were several ethical considerations that were dealt
with. The interventions were outlined and considered by the university ethics committee
to determine appropriateness Approval was given by the committee for the
interventions to be carried out. The major consideration revolved around confidentiality
and was dealt with by careful collection storing, and disposal of infom1ation. A second

5.1

consideration was the sensitive nature of the subject matter To deal with this
consideration debriefing occurred at the end of each of the interventions and at pre and
post test students were assured that they could acces'; the Counselling Centre and/or the
researcher at any time shouJd they feel they needed further debriefing.

Data Analysis
Data analy\is was conducted in three areas, analysis of the items of the
questionnaires. analysis of the Likert type scales in the attitude and hehaviour
questionnaires, and exploration of responses to the demographic questions and the
open-ended questions in the attitude and hehaviour scales.
Because the constn1ction of the items in the questionnaire package was
developed by the researcher, and not subJected to tests for external validity, an item
analyses for both the attitude and the behaviour questionnaires was necessary. An item
analysis gives the researcher information about whether response alternatives function
welJ for an item, if an item is measuring appropriately and if each item is contributing to
the reliability of the questionnaire (Assessment Systems Corporation, 1993 : Sax, 1989).
Due to the variance of questioning and the demographic nature of Questionnaire # I, an
item analysis was not done for this portion of the package. The demographic
questionnaire had been part of the piloting process mentioned earlier. The item analysis
for both the attitude and behaviour scales was peribrrned with the use of the computer
program ITEMAN1m (Assessment Systems Corporation, 1993).
ITEMAN1t11 offers a series of choices tor the configuration of analysis and

information presentation once the analysis is done. The item statistics for multi point

54

items provided by ITEMAN1111 included the number of examinees endorsing each item~
mean subject !)Core, the variance of distnbution of subJect scores, standard deviation of
distribution~skewness of distribution. minimum, median, and maximum scores. mean
Pearson correlation~ and, several other re1:>ults which were not all useful for the
researcher's purposes The main statistical process of concern in this item analys15 was
the item scale correlation by which responses to a particular item were correlated with
the average scores for exa1ninees. \Vhile other statistics provided useful information, it
was the item scale correlation that was of prime consideration in the results of the item
analysis The correlation used was a Pearson product-moment correlation between an
item with a possible range of I to 5, or I to 6. and the total score with a range of
minimum to maximum.
To deal with missing data, the researcher chose to endorse itemwise deletion.
This process allowed for all subject information although they may not have responded
to all questions Therefore, there was potential variation in number of examinees within
each scale. This variation was chosen due to the fact that it had less influence on the
outcome of the analysis than a deletion of respondents that did not respond to all
questions, as would have occurred with scalewise and casewise deletion.
The Analysis of Variance (ANOV A) was used to investigate the potential of
significant change in attitudes and behaviours caused hy the interventions. The
ANOVA is a statistical procedure involving the comparisons of means to determme if
the differences between these two means arc more than could he attributed to

55
sampling error (Glass & Hopkins, 1984). While the necessary analysis could have
been done through separate t-tests and paired sample t-tests, the ANOV A was seen as
a 1nore accurate and powerful test. Glass and Hopkins pointed to three advantages
that the ANOV A had over t-test\:
( 1) It yields an accurate and known type-I error prohahility, whereas
the actual a for the set of <ieparatc t-tests is high yet undetermme<l;
(2) It is more powerful (when a is held constant)-that is, if the null
hypothesis is false, it is more likely to be rejected:
(3) It can assess the effects of two or more independent variables
simultaneously (p 32S)
One-way ANOVA is used when groups are defined on only one independent variahle
(Howell, 1995). In this case, the independent variable was the intervention or lack of
intervention given. A preliminary one-way ANOV A was done to determine any
interaction of gender. The results predicted the type of ANOVA done for the
analysis. No significant difference was found for gender, therefore, a one-way
ANOV A was decided upon for the analysis. A second set of one way ANOV A were
done on the attitude and behaviour scales to determine any significant difference
between the groups at the pre test level. Jf there was no significant difference. gain
scores could then he obtained and another set of ANOVA could he done to see if
there was any significant change in the groups. Wiersma ( I986) reported that there

56

are those who have cautioned against the use of gain scores in statistical procedures
due to the believed limitations. Wier\ma asserted that gain scores were desirahle in
certain areas, particularly in instructional area\, and presented opinion, of other
authors who expressed that gain scores could he reliable and useful.
A test of multiple comparisons would he done for any of the I\ NOV A if the
null hypothesis was rejected. With rl!Jection of the null hypothesi~ detection of
significant differences in mean~ ts important (Glass & Hopkms, 1984). Multiple
comparison techniques are used for "making comparisons between two or more group
means subsequent to an analysis of variance" (Howell, I995). When sample groups
are equal, as in this case, Tukey's HSD test provides an easily understood and

interpreted process for investigation of the groups presenting significant difference.
The minimum difference between means necessary to r~ject the null hypothesis is
computed, this is the honestly significant difference. The two means that difter more
than the honestly significant difference are said to be significant. Glass and Hopkins
(1984) suggested that the chance of type one error occmTing in the contrast~
(differences between means) is lower than for other tests of multiple comparison,
although power is less with this type of proces\ and so more of a chance of Type 11
error. The Tu key tes1 for Multiple Comparisons is a conservative test. Other tests of
multiple comparisons provided for Jess conservative results and, thus, less chance of
Type II error. Two such tests inclu<..led the Neuman ·Keuls and the test of multiplet'~

'i7

or Least Significant Difference (LSD). The Neuman Kculs was presented by Glass
and Hopkins (1984) as very similar to the Tukey test hut with more power. That is~
this test was suggested to r~ject more null hypotheses than Tukey's test. Following an
ANOV /\, multiple t-te'\tS can he done to determine the means that presented

significant difference. These te~ts arc the least conservative of the multiple
comparison test\, therctore the chance of Type II error is further dealt with.
However, Glass and Hopkins presented the argument that it was very difficult to
determine whether Type I error had occurred or not with LSD.
With a failure to r~ject the null hypothesis stati\tical evaluation ends. The

alpha level, or probability for Type I error, for all ANOVA 's done wa\ a = .05; the

most common size of rejection region ( Glass & Hopkins, 1984; Howell, 1995).
The researcher explored the demographic and open response data through a
process of scanning. The questions were explored for themes and suhthemes hy the
researcher and an assistant, agreement hetween the two reviewers. In this way themes
that were observed in the post test but not in the pre test, and themes that appeard in

greater frequency at the post test level were recorded.

Summary
This chapter presented an overview of the process of the study including
subJect recruitment, item construction, the procedure used in the administration of the
pre and post tests and interventions, and the planned process of analysis of su~jcct

58

responses. In Chapter 4, the researcher describes the item and data analyses and
discusses the results of these analyses. An exploration of the open-ended response
analyses and the results is also included in this chapter.

59
CHAPTER4

RESULTS

Chapter 4 includes the results of all analysis done on the data obtained from the
questionnaires completed by subjects pre and post intervention. The outcomes of an
exploration of demographic responses is presented as well as results from the item
analysis The results of analysis of both the closed and open-ended responses of both the
attitude and behaviour questionnaires are presented. A discussion of the results and their
meaning in the study completes the ResuJt5 section.

Discussion of Demographic Data
The mean age of participants was 20.6, with very little diflerence between
groups, ie. Control Mean = 20.8, Intervention I Mean = 20 5, Intervention 2 Mean =
19.9. The researcher also explored the modes of all groups and each individual group
Overall the age of participants was bimodal~ 18 and 19 yrs.
Subjects were asked about their ethnic/cultural background. Twenty-five of the
31 subjects (80%) defined themselves as Canadian. One person was identified as Irish,
another person identified themself as Chinese, and a third identified their culture as
Norwegian, a fourth was Brazilian, a fifth Portuguese, and a final subject identified that
they were British. Of the group who identified themselves as Canadian, most identified a
european heritage, particularly Scottish (6), Irish (7), and English (8). There was one
subject of the group of 31 who particularly identified that they were a Newfoundlander
Five out of the 25 students (20°/o) who identified themselves as canadian also identified

60
as being caucasian. with 4 out of the five identifying that themselves as white males
The ratio between males and fomales participating in the study was 12 males to
19 females There were 80% or the subject group who identified themselves as strictly

heterosexual. Four out of thirty-one ( 13%) rated themselves as a two on a scale of one
to six from ··strictly heterosexual" (I) to ''strictly homosexual" (6). One subject rated
self as strictly homosexual. The amount of post secondary experience of the subjects
involved in the study ranged from first year to graduate work. The mean for years of
post secondary study was found to be 2.2 years, using a 5 to represent the 4+ category.
The most frequently occurring respom,e.., presented as bimodal with first year being the
response for IO subjects and second year being the response for IO further subjects The
majority of subjects, 19 (61 o/o), reported having some information about HlV provided
for them in high school The most frequent types of information given by respondents
were such things as information about contraction, high risk behaviour, and condom use
Twenty-four out of the 3 l subJects (77%)reported speaking to their families very
rarely, or occasionaJly about HIV The family member most mentioned as the person to
whom they had spoken about this topic was their mother. Twenty-nine of the subjects
(93%) reported speaking to friends about HIV. When asked if they communicate with a
partner around HIV, 26 of the subjects said that they did, the others responded with a
"no" or "not applicable" Of those who responded with a negative, several made the
qualification that they have had no sexual experience and thus the question would not
necessarily be applicable to them. The subjects were asked if they knew anyone who ~as
HIV +. In this question 20 out of the 31 subjects (64%) responded with a negative.

61

Eight of the 31 subjects, about 26o/o, said that they had been tested for J IIV in the last six
months Most subjects reported having learned about HIV from the media, high school,
parents and friends . It is interesting that of the 31 subjects involved in this study, 8, or
26%, of them reported that they had not yet had sexual intercourse

When questioned about whether subJects felt that it was important for
them to learn more about HIV, 90 % responded aflirmat1vely. The most common
reasons given for this response reflected personal risk and the necessity to keep up on
infonnation to decrease persona] risk. Although termed in a variety of ways, I 7 subjects
addressed the importance of education in prevention of HIV. Another common response
revolved around acknowledgement of danger in others. Several subjects responded by
saying that they could also help others to protect themselves. Other subjects
acknowledged that they did not know enough about HIV. Two subjects responded with
a "no" to this question One subject appeared to be basing some of this decision on
having a homosexual orientation. "rm not worried, although J probably should be ... but
everyone downplays the possibility of lesbians contracting''. The second subject revealed
making lifestyle decisions in which contraction was not a problem as the major reason for
their belief that they did not need to learn more about HIV.
In Table I, a summary of responses to the demographic questionnaire is
presented. Type refers to the types of questions asked on the questionnaire, ( ic. I am
_ Male _ Female) and number of subjects refers to the number of subJects tor whom a
particular type was reflective. Mode is used to reflect the responses given by the
majority of the sample group

62
Table I

Summan of Demographic Data

-

Type
~

-

No of Subjects

- - - - - - - - - - - - - - - - ---

Age-Control
Age-Intervention I
Age-Intervention 2
Gender
Male
Female

Male-Control
Female-( ontrol
Male-lnten·ent,on l
Female-lntencnt1on l
~fole-lntcnenllon 2
Femalc-lntervcnt1on 2
Sexual Orientation

12
19

Mean

Mode

70 6
20 8

18 & 19
24

20.5

20

I 9.9

19

FcmnJc

3

<,
5

Female
Male & rcmalc

5
4

8

strict!) hctcrosc,ual
slnclly homosexual

Female

Strictly l lctcro

other
HIV Info. Prondcd in Highschool

yes

19
12

no or no response
Communication with Famity
Commu01cat10n \\ 1th Friends
Yes

29

No

2

Commw11caL1on with Partner
Yes
No/NA

Yes
Rare I\

26
.5

Yes

Yes

Contact with PLWH
Yes

No

II
20

No

8
23

No

Tested in Last 6 Months
Yes
No

Learning Aboul HIV 1s Important
Yes

No

28
2

Yes

.
Note, lntcrvent1on I - lntcrvcnt10n Group I~lntcn cnlton 2 = l11tcr\'cnllon Group 2, PLWH =

Persons living \\1th IIIV

63

Presentation of Results
Item Analysis
In the Item analysis two considerations became part of the decision to accept or
reject an item that did not have an acceptable correlation The number of subjects rn this
study was 31, therefore, an item with a correlation below - 3 I wou ld not be acceptable
as it demonstrated a significantly (p "'- 05) negative correlation with the rest of the
instrument The other consideration for deciding to retain items was the value of
inclusion of that particular item in the data analysis. The item analysis of the attitude
questionnaire presented acceptable correlations for all items.

In the behaviour scale, several questions were explored to decide whether to
retain them. All questions where the correlat1on was < -.30 were explored Question

# 12, r = - 53, was omitted from the analysis. This question dealt with personal
responsibility in provision of condoms/ protection . While the question was considered to
be valuable, disparity between acceptable con-elation and r value for this particular item
was too great to include the item in the analysis Question # 14, r = - 28, while within
acceptable range, dealt with needle sharing which was not considered to be a major
concern and, therefore was omitted Question #15, r = - 36, and Question # 17, r = -.37,
dealt with alcohol influenced behaviour and response to coercion respectively. The
researcher made the decision to retain these two items due to the importance of the items
to the scale and due to the relatively small differential between correlationaJ scores and
the acceptable limit - .30. Questions# IO r = -.27, and# 19 r = -.29, were considered as
well due to t heir proximity to -.30 Both questions dealt with the issue of multiple

64
partners The questions were reflective of each other, therefore, the researcher made the
assumption that the problems 'with these questions were either a result of the wording or
the single concept covered by both questions. Questions # 10 and # 19 covered a very

important concept to I IIV prevention
In Table I[, the researcher provides a summary of some of the information from
the data analysis
Tahle LI

Summary Table for Item Analysis of Questionnaires

--

Source

No. Items

No. Examinees

Mean

Std Dev.

Alpha

SEM

Attitude
Behaviour

32
31

31
31

2.1
2.8

0.4
0.4

.7
.7

0.2
0.2

Note: Demographic Survey was not included in Item Analysis;
Alpha = Cronhach's Alpha; SEM = Standard Error of Measurement
For both scales the standard error of measurement was 0.2. The closer the
standard error of measurement is to 0, the less variance between the true and ohtained
scores (Sax, 1989). The standard error of measurement is interpreted as a
standardized deviation for measurement error. In this case a person who ohtained a
score of 2.1 would be expected to ohtain mean scores of I .9 to 2 .3 68% of the time.
For the behaviour scale, the standard error of measurement was the same as that of
the attitude scale, with a mean of 2.8 there could be a variance from 2.6 to 3.0. With
this range of 0.2, the rcliahility of the scale is judged acceptable for their intended

65

purposes.

Close Ended Responses
An ANO\'A was can-ied out to test for possible attitude difference among the
three groups prior to the treatment. No significant difference was found . Summary
ANOVA results are given in Table JI I.

Table III

One Way Analysis of Variance for Attitude - Pre Test
Source

df

F-Ratio

p

Between Groups
Within Groups
Error

2
28

0.69

0.51

~

(42.73)

- Value enclosed in parentheses represents mean square error: df = degrees of

freedom *p < 05 .
As the F ratio is less than 1, it is clear that there is no significant diflerence
between the groups in the attitude questionnaire The procedure was repeated for the
Behaviour questionnaire Results are given in Table IV.
Table IV

One-Way Analysis of Variance for Behaviour - Pre Test
Source

df

F-Ratio

F-Prob.

Between Groups
Within Groups

2

1.41

0.26

Error

28
(368 64)

- - - ----

---

~
- Value enclosed in parentheses represents mean square error~df - degrees of
freedom .* p < .05

66
According to the results of the ANOV A for the behaviour questionnaire, the
F-ratio was not sufficient to reject the nuJI hypothesis. While the ratio was above 1,
it did not approach the critical F; F Ratio : 1.4 l < F criticaJ: 3.34.

Results of the above A NOVA showed no significant difference among the
three study groups as measured on either the Attitude or Behaviour tests. Two more

One Factor A NOVA were performed to determine any significant change within
groups pre and post intervention. The ANOVA was executed using gain scores
determined as a difference post scores - pre scores Results of the One Factor
ANOVA are presented in Tables V and VI.
Table V

One Way ANOVA with Attitude GAfN scores
Source

df

F-Ratio

F-Prob

Between Groups

2
28

83

.45

Within Groups

Error

(394 65)

Note Value enclosed in parentheses represents mean square error: df = degrees of

freedom. * p < .05
No significant difference was found in attitude among the groups. As the
critical F (2,28) is 3.34 and the R ratio is .83; the null hypothesis is retained.

67

Table VI

One Way ANO\'A with Behaviour GAIN scores
- - - - - - - - - - - - ~ -------- -- --~
Source

df

Between Groups
Within Groups
Error

2

F-Ratio

F-Prob

- --------------~
4.44*

.02

28
(394.65)

Note Value enclosed in parentheses represents mean square error: df =- degrees of
freedom . * J2 · OS
Significant difference was found in the hchaviour questionnaire. A Tukey·s
honestly significant difference multiple comparisons tc~t \hows the areas in which
significant difference was found. There was a significant difference in means hetween
Intervention 1 group, M = -2. 9 and Control group, M = 21. Due to the increased
chance of Type II error is using the Tukey Test of Honestly Significant Difference
(Glass & Hopkins, 1984), hoth the Neuman-Keuls and LSD were also run. Both of
these tests for multiple comparisons showed the same result<, as the Tu key.

Open Ended Respons~_s
The open ended questions provided an opportunity for the researcher to explore
responses that may not have occurred in the Liker type scale. In this section, major
themes occurring in the responses are presented. The relevance of the themes will be
addressed in the next section A general theme that occurred in all three groups was a
decrease of information given by respondents. In the post test. respondents appeared to
feed back more detailed intbrmation in their responses.

68

Attitude Questionnaire. In the attitude questionnaire, the questions explored
such areas as, behaviours seen as preventive, importance of safer sex, beliefs about
groups who contract HIV, opinions about abstinence as an option, perceived barriers to
safer sex, and perceptions of group opinion on safer sex. At the pre test level, all three
groups had reflected the use of condoms as the maJor method of preventive action
There was also a focus on knowing one's partner. Abstinence was presented as an
option, but one that v. as unrealistic. Little change was observed in responses of those in
the Control group between pre and post test, with the exception that more subjects
expressed the opinion that it was important for people to accept responsibility for their
actions. This change was also reflected in the post tests of the other groups. It was
observed. however, that responses for Intervention 1 expressed more concern about
interaction and protection of self The subjects presented communication with partner as
a high priority, rather than merely knowing one' s partner. Subjects m this group also
expressed concern about the fact that they could not identify someone who was HIV
positive at the post test level, this was not reported in the pre test by any groups
Subjects in Intervention 2 were more information oriented and more specific Il was in
this group that such information as using spem1icides such as non-oxyl 9 were found . A
further change from pre to post test responses in both Intervention I and Intervention 2,
although stronger in Intervention 2, was that subjects were less inclined to express
abstinence as an unrealistic option. The shift seemed to be from "nice, but it won ' t
work" to "it is a reasonable choice that may be taken by some people". In Intervention 2
there was a concern expressed that the promotion of abstinence as a viable option would

69
deter information being given about safer sex. This concern had not been mentioned in
the pre test
There was little to no change in responses at the pre and post levels regarding the
opinion that it was important to practice safer sex. The vast majority of subject<; from all
groups agreed that this was important. Those who did not agree with this opinion did so
because they were (a) in a monogamous relationship, (b) had not yet had sexual
intercourse (c) proported to use condoms every time they had sex One minor shift in
responses occurred with Intervention I where the respondents expressed more caution
around the beJief that knowing their partners would protect them from HIV contraction.
All groups responded that they thought anyone could contract I IIV. They also
displayed kno\vledge of the groups that were considered to be high risk. The only
change observed at the post test level was in Intervention l where some subjects focused
on children and the potential for children to contract in vitro and from high risk parents.
All subjects at the pre and post levels seemed to respond in similar fashion to the
types of barriers to safer sex. They presented irresponsibility, need to purchase
condoms, lack of knowledge, embarrassment. lack of communication with partner, use
of alcohol, perceived invulnerability, and peer pressure as barriers lo preventive
behaviours. Most respondents presented the belief that their peers generally supported
the use of condoms and practice of safer sex, but were not practising the behaviours.
Behaviour Questionnaire. In the behaviour scale, the open-ended questions
focused around communication with sexual partners, situation in which unprotected sex
occurred, types of safer sex practices used . There was one question on the use of

70

intravenous drugs. No respondents admitted to having used intravenous drugs. Some of
the responses changed between pre and post test also because sexual behaviour was not
measured for as long a period of time at the post test When sub1ects \vere asked if they
commurucated v.ith their partners about safer sex, the majority in both pre and post tests
responded affirmatively, again several of the no responses were a result of lack of sexual
interaction. As was reported in the demographic discussion, eight out of the 3 l subjects
reported that they had never had sexual intercourse. One group, Intervention I, showed
a great difference between pre and post test responses At the pre test level, five
subjects reported that they had not commurucated with partnerc; around safer sex, and
seven said they had In the post test. only one person responded negatively to the
question and Intervention 2 reported that they had spoken to partners about safer sex.
The ways that subjects communicated were similar at pre and post levels, however, at
the post level subjects in Intervention I gave more examples of the ways they
communicated and were more detailed in their examples For all groups at the pre and
post test levels, the main situations in wluch subjects had engaged in unprotected sex
were when they had been drinking, when they "got caught up" and didn't want to stop
because there wasn't a condom available, or when they were in monogamous
relationships. Both genders reported that they had engaged in unprotected sex due to
pressure from their partner, although there was a higher concentration of women with
this response Using condoms was the most common rci;ponse to the types of safer sex
practice, although again several respondents in the various groups responded that they
were abstinent with three qualifying that they intended to stay that way until marriage.

71

In Intervention I and Intervention 2, more respondents began to look at being tested for
HIV as something that they should do Respondents in Intervention 2 also reported, in
the post test. that they were using spermicides with condoms. Generally there was a
theme of increased caution around sexual activity in the responses to the post test. This
theme was seen in all groups includmg the control.
Post Test Addition. An e.x1.ra question was asked at the post test to explore
subjects' ideas of their learning from the interventions. One subject from the Control
group responded that they had been impacted by the questionnaire and realized that they
needed to discuss sexual history with their partner. Only one subject responded that they
felt they had not learned anything from the intervention they had attended (Intervention
1). However, this subject reflected enjoyment of listen mg to the guest speaker and that

they were helped in seeing the relevance ofHlV prevention to self It is wondered if the
subject actually learned nothing from the intervention. Very different themes were found
in the responses of subjects from Intervention I and intervention 2 to this question.

Subjects in Intervention l appeared to respond in a more emotional manner than those in
Intervention 2. They reflected some of the specific HIV infom1ation they had heard and
acknowledged their own lack of knowledge in certain areas The subjects reported being
moved emotionally by the treatment of the presenter by other people since she has been
l-IlV positive - they expressed anger at those who had treated this person m a negative
way. Several of the subjects reflected that they felt a deeper respect for persons living
with HIV and their ''bravery" . One subject reflected that this aflective intervention had

given them a "wake-up caH". that they needed to change their nsky beha\ tour. Most of

72
the members of this group reflected that the intervention had helped thern reaJly
understand that one could not tell a person was HIV positive by looking at them One
subject put it ~ell when they said. "I always knew that you couldn't tell when someone
had HI V, but 1 thought that I would still know". One client commented on the
usefulness of the goal setting, and another spoke of coming Lo the conclusion that
practising safer sex was "easy". t\ tlirther theme was that these subjects did not want to
have to experience what this woman had experienced. and expressed a desire to change
what they were doing to help themselves avoid this risk One subJect added that they
were helped to be aware that there was "life after HIV", which, in turn, helped them
understand that HIV was not to be feared, but to be prevented.
Subjects who had attended Intervention 2 reflected a greater concern for others
and for the part they could play in helping people increase thetr knowledge about HIV
and in protecting themselves They were also able to indicate the new knowledge that
they had gathered through this intervention These subjects also said that they became
aware they did not know as much about HIV as they thought they should. They also fed
back that accurate knowledge was one very ma.1or factor in preventing HIV. Two
subjects referred specifically to wanting to become involved in educating in the area of
HIV and many reflected a need to help their friends and people around them to become
more informed about HIV. A maJor theme m these responses was personal
responsibility. These subjects acknowledged that there were barriers to preventive
behaviours, but that in most situations people had some control over their behaviours
Three of the respondents wrote about the importance of HIV testing and two reflected

73

that they intended to be tested . One interesting response was that a subject had learned
about the potentially different ways that men and women had to regulate sex and
protection They also commented that they had become more aware of this factor in
their own relationship.

Discussion of ~tlU
In this section of Chapter 4 discussion occurs around the results of the item
analysis, data analysis and scanning done on the open ended questions. An item analysis
showed that both the attitude and behaviour scales were reliable. This result allowed for
a continuation to the data analysis with only minor changes in the behaviour
questionnaire and no changes in the attitude questionnaire Any measured lack of
difference among groups cannot be attributed to the unreliability of the instruments. The
items were shown to be measuring what they were designed to measure which increases
confidence in the results of the data analysis The potential to use this questionnaire
package with some alterations 1s supported by the results of the item analysis.
The results of the first set of ANO VA for attitude and behaviour scales (see
Tables III and IV) at the pre test level showed no significant difference between the three
groups. This meant that the two intervention groups and the control were all considered
to be equivalent before the interventions took place. The result allowed the researcher
to follow through with analysis of gain scores to establish any difference from pre to post
test between the groups.
The ANOVA, done on gain scores for the attitude quei;t1onnaire, did not measure

significant differences between any of the groups. However, investigation of the open

74
ended responses did show some movement. Subjects who had attended the affective
intervention (Intervention l) showed an increase in caution around trusting tha1 partners
did not have HIV. This result could well have come from the woman's relaying that
neither she nor her partner had known that he was I IJV positive. The attitude that
knowledge of one' s partner being a primary method of mv prevention appeared to shift
for these subjects. This shift was not seen for the group who received the second
intervention. Subjects in Intervention I also began to deal with the fact that they could
not distinguish between people who were HIV positive and those who weren 't. This led
them to stress the attitude that anyone could contract HJV Subjects in intervention 2
reflected more concern about personal responsibility and expressed the belief that people
did have some control over their sexual behaviours. This trend was not seen in the pre
test. Subjects in this group also expressed more concern about helping others to educate
and protect themselves While these changes were seen, it is quite possible that they did
not occur in amounts that could be measured as statistically significant.
Significant attitude change may not have been measured due to the limited
sample size With a sample of 31, people it is difficult to attain a statistically significant
score. For this reason, alpha levels for ANOVA's with smaller groups can go as high as
.10 (Glass & Hopkins, 1984) The alpha leveJ used here was .05 to eliminate increased

risk of Type 1 error. Another possibility is that the Likert scales did not measure all of
the attitudes and beliefs impacted by the interventions. For example, while the items did
measure beliefs about the need to communicate with and influence significant others,
there was no measure for amount of concern or feelings of responsibility for the welfare

75

of others This was one of the stronger themes in the open ended responses in the post
test, particularly in Intervention 2 A more detailed exploration of the ways that subjects
responded to the Likert scale and closer comparison to the themes in the open ended
questions may help in exploration of the insufficiently significant result
Results from the ANOVA done on gain scores for the behaviour scale showed
significant difference. The two means with significant difference were the mean for
control group at -2.9 and the mean for Intervention I at 21. This resuJt pointed to
positive change in behaviours of those subjects who had been involved with the affective
intervention, Intervention I. It would appear that an intervention exposing subjects to
someone who has HIV, that also accesses emotions, is potentially effective in promoting
positive attitude change. Open ended questions responded to by subjects in Intervention
I, also provided information to the researcher about change in behaviour that occurred.
There was a strong change between pre and post test around communication with
partners At the pre test level seven subjects reported that they communicated with their
partners about HIV, while at the post test level, 12 responded aflinnatively to this
question. There is a possibility that the process of being involved in the intervention
itself and thoughts about the intervention jusl after it occuffed promoted discussion
among partners. More detailed investigation of individual questions in the behaviour
scale and change in these questions pre and post test may aid in determination of the
exact behaviours and how they were impacted by the interventions.

Summary
In this chapter responses to demographic items were explored Results of item

76

and data analyses were reported and responses to open ended questions of both attitude
and behaviour were investigated Discussion of the responses to the questionnaires
concluded the chapter In Chapter 5 the study will be discus~cd in more general terms
Limitations will be explored and suggestions made tor further research in this area

77
Cl IAPTPR 5
DISCUSSION

In Chapter 5, the study is di~cussed in general terms Implications of the results
of the study and conclusions drawn are highlighted. Limitations to the research
are explored and ideas for fur1her research in the area of I IIV prevention are suggested.
General Discussion
This study involved the development of two interventions designed to promote
positive attitudes and behaviours relevant to HIV prevention. Information that provided
an added dimension to the overall interpretation of the study results was found in the
open-ended responses to the attiLude and behaviour scales post test. An analysis of the
close ended responses showed a significant change in behaviour for those subjects who
attended Intervention J.
Hypothesis I
The first hypothesis stated that an intervention in which subjects were presented
with accurate HIV information, given the opportunity to interact with a peer who was

HIV positive, and provided with the opportunity to set goals around preventive
behaviour would promote positive attitudes and behaviours. The first intervention tested
this hypothesis A significant difference in behaviour pre and post test supported the
hypothesis
Subjects can be aided in changing behaviour through modelling (Bandura, 1977)
and the first intervention was based on the premise that people can leam vicariously.

78

Cotton et al. ( 1994) stressed the importance of peers in persuasion They pointed to the
efficacy of print material relating stories of those who are living with HJV. This
researcher hypothesized that an intervention in which a person living with ll lV was
present and relating her story would also be an effective method to promote positive
attitudes and behaviours. The speaker was both a peer of the participants and a woman
who had contracted HIV through heterosexual contact Cohen and Bruce ( 1997)
reported that people have a tendency to believe HIV risk to be stronger for peers than
for self This may have been a factor in lack of change tn attitudes
Sandfort and Zessen ( 1992) stated that interaction of people with someone who
is living with HIV can decrease the tendency toward denial of personal susceptibility.
The hypothesis that an intervention exposmg subjects to a person living with IHV would
be effective was supported by a statistically sigmficant change on the behaviour scale
completed by subjects.
Intervention I could bring several very powerful messages to educative
programs.
I . Peers do contract Hl V.
2. It is not possible to tel1 that someone has HIV by looking at them or
through "scoping them out".
3. HIV is not to be feared but to be addressed and precautions should be

taken to reduce risk.
4 . Groups are not risk factors, behaviours are. The current study suggested

that interaction of young heterosexuals with a peer, who is HIV

79

positive, may add to the understanding that the threat of AfDS
comes less from being a member of a particular group and more from
engaging in risky behaviours.
This intervention could be a very powerful element in a prevention program
The amount of emotion expressed by the subJects in the intervention attests to its

power They expressed anger at the way persons livmg with HIV are treated The
subjects also reflected a growing sense of respect for people who are HIV positive. An
interesting reaction that occurred in open-ended responses of these subjects was a
highlighting of awareness of the interaction between pregnancy and high risk behaviour.
Many subjects responded quite emot1onally to the risk encountered by the woman's
children due to her contraction while she was pregnant. It would be interesting to
explore whether this consideration would be an added factor in deciding to practice safer
sex, and whether this decision is different according to gender
Hypothesis 2
An analysis of data taken from those subjects who attended the second

intervention showed that no statistically significant change occurred between the three
groups in either attitude or behaviour. In this case the null hypothesis could not be
rejected and so hypothesis 2, subjects who were engaged in the development of
preventive strategies for their age group would tend to develop more positive attitudes
about and behaviours toward HIV, could not be supported

The open ended questions

asked in the questionnaire package did reveal some interesting responses in the post test
that had not occurred in the pre test. There is a possibility that the destgn of this

80

intervention rather than the premise behind the design presented the difficulty in
achieving change The Centers for Disease Control ( 1996) supported the use of peer
educators in intervention activities. This organization reflected that th1s method
11

provides an opportunity for individuals to perceive themscl\es as empowered by helping
persons in their communities and social networks, thus supporting their own health
enhancing practices" {p. 3). The subJects were only involved in a very preliminary stage
of intervention development which may not have allowed for time to begin to associate
with the attitude object Due to the time limitations, subjects would very likely not have
felt any sense of ownership for the interventions they designed and so there would be
less threat to them if their attitudes or behaviours did not change to reflect their
intervention. While the second hypothesis was not supported in statistical analysis, this
intervention is still seen as potentially very valuable Intervention 2 is worthy of further
exploration with extended time allowing for complete design and implementation of the
interventions by the subjects
Responses to open ended questions for both questionnaires pointed to the need
for an educative component in interventions. While accurate knowledge is not sufficient
to promote attitude and behaviour change, it is a very important element. In the post
test, subjects from both intervention groups showed increased knowledge about certain
aspects ofHIV prevention. For example, subjects in Intervention 2 began to identify
spermicides as a necessary part of the process of using a condom Any mten. ention for
health prevention must have an educational component in whkh up to date and accurate
information is taught.

81

Interventions 1 and 2 shared two advantages for use alone or within a prevention
program. The first advantage is that the interventions are versatile They can be used
for small group work. in larger groups, or on an individual basis. Because of the many
challenges to JIIV prevention, such as designing interventions that may be effective in
several areas with several groups, versatility of interventions is of prime importance.
While individual and group differences must be considered in any intervention, it may be
time to explore common factors, established through structured studies, in risk
behaviour for HIV to determine elements that can be used across programs and
interventions.
A second advantage of both interventions 1s that a small number of resources are
necessary for implementation. The information portion of both interventions was based
on relevant literature made available, to the author, by the National AIDS Clearinghouse
Through this resource, tapes, posters, articles, books and full preventive program
information was accessed. Quantity of materials was suflicient for all groups in
Intervention 2 to complete their intervention planning. The only other requirements for
these interventions was an overhead and the use of a large sized room. The young
woman, who spoke to subjects about her experiences, did so on a voluntary basis. With
limitations to funding it is presumed that interventions which requrrc little monetary
involvement would be very helpful.

Limitations
As in any study dealing with human attitudes and hehaviours, there are factors
that present limitations. The number of factors that contribute to the way a person

82
wilJ think, feel and hehave are extensive and very difficult to measure. ractors such
as the influence of the larger system. the situations in which certain affectivc, mental
or behavioural element5 occur, and view of c;;elf all have an impact on how a person

will think, feel and hehave. Re~ults rn stu<lie\ such a, the one presented in this thesi\
must he interpreted with the acknowledgement that the vast array of contrihuting
factors to human functioning cannot all he controlled in any cxpedmcntal situation.
One limitation to th1, s1udy involved the potential difficulties ot uc;;e of verhal
expression to measure change m behaviour. The inahility to u,e ohservationaJ
methods to determine hehaviour change (Sorenson et al., 1991) is a limitation . For
example, a possible interaction would be the need for self presentation (Catania et al.,
1993). Ostrow and Kessler ( 1993) suggested that suhjects' responses may not be fully
representative of their behaviours due to need for privacy, <;elf enhancement and
embarrassment. Subjects may have reported attitudes and/or behaviours that presented
themselves in a positive light, as defined by those su~jects. A related limitation
would involve the subjects presenting attitudes and/or behaviours for which they
expected the researcher to be lookmg. In hoth of these cases, the attitudes and
behaviours measured may have reflected more of an attempt to present the \U~ject in
a positive light or to please the researcher hy responding in a way m which they felt
the researcher wanted. If self presentation or responding to the researcher were
occurring to any significant extent, it should have been reflected in all three groups at

83

the post test level. There was also an attempt to deal with social response hias, both
through instructions made hy the researcher to the suhjects that responding to the
researcher would not he helpful, and through the amount of effort that went into
maintenance of confidentiality.
A second limitation of the c;tudy was numher of subjects (N = 31) . This was
a small study. therefore , the result~ would he more difficult to generalize to the larger

population (all student<; 18 through 25 at the University of Northern British
Columbia). With a low sample number, there is less likelihood that all members with
different characteristics are represented. One way to investigate this would he to have

more socio-demographic characteristics measured in the demographic information.
The larger the number of subjects from a certain population involved in the study, the
greater the chance that major group characteristics are included and the greater the
power of the study (Howell , 1995).
Subjeclc;' behaviour were measured over six months in the pre test and only 7
weeks in the post test. This must he considered a possible limitation to the study .
Sexual behaviours are related to many various factors such as involvement with a
partner at a particular time, amount of time spent in sexual activity and opportunity
for sexual activity. Behaviour, measured over a six month time period, may have a
marked difference to behaviour measured over a 4ievcn week period. A period of ~ix
months would appear to present opportunity for more sexual hehaviours to he

84

measured than a smaller period of time. However, at the pre test level, all groups
were presumed to be equal according to the results of the two ANOVA done. If there
had been a difficulty with the amount or type of sexual behaviour measured, due lo
time differences, it should have been seen in all three groups in the analysi~ for the
post test. This was not the case and only Intervention 1 presented marked change in
behaviours hetween the pre and post tests.
While preliminary behaviours showed significant change in Intervention I,
there was no f oJlow-up to check relapse of behaviours. It is quite possible that the
initial response to the intervention, particularly an affective intervention, dropped off
quickly as the suh.1ects moved back to habitual ways of behaving. To he able to state
that behaviour did change in more than a preliminary fashion, a follow-up study
would need to he done, to measure behaviour after an increased amount of time had
passed. The researcher would pr~ject that the interventions may he more successful
in behaviour maintenance with the inclusion of practice in sexual negotiation and
condom use. Sexual negotiation was addressed in the information session and
examples of successful methods of negotiation were presented. However, the subjects
did not have the opportunity to practice these methods.
A further limitation to this study was the mstruments used to measure change.
The questionnaires were de~igned by the researcher an<l this study was the first in
which they were used. External validity is increased when replication occurs.

85

Therefore, the questionnaires would need to be replicated with simi lar results before
the researcher could begin to propose that the questionnaire was externally vaJjd. Tp
a great extent. the piloting process involved feedback from those on whom the
package wa~ piloted. This process may have heen improved by more statistical
exploration through an item analysis before the ~tudy was carried out. Item analy~is
would a11ow the researcher to make changes in the questions before they were used

with the sample group in the main study. In this way, the amount of questions

omitted from data analysis may be lowered.
The forms of sampling used in this study were convenience or accidental

sampling and snowballing It is suggested that, although conventence sampling is used

extensively, there is a difficulty in generalizability in those studies m which convenience
sampling occurs. Convenience sampling is not considered to be random sampling Glass

and Hopkins ( 1984) suggested that conclusions, made in studies in which sampling
occurred, should be made with caution. Like the lack of sample size, this form of
sampling can lead to an inability to generalize to the population. With convenience
sampling, there is a possibility that the sample will not represent the parameters found in
the population and, so, the results of any study could not be generalized to the

population from which the sample was taken.
A final limitation is also related to the sampling used in this study Participation

in the study was strictly voluntary with no reward system in place for those who took
part Sampling was done through both convenience and snowballing techniques. The

86

subjects would be assumed to have some type of motivation to volunteer, they had either
been pressured by friends to take part, or would have an interest in the area of HIV
With an increased level of motivation, the results of the study may also reflect a factor

outside of the considerations of the study. Subjects were \Cry likeJy motivated to al least
learn more about HIV and preventive strategies, which could lead to results that would
not necessarily be seen in samples that were gathered through random selection.

Suggestions for Further Research
This study resulted in the formulations of many more questions than were answered
In this section the researcher will explore some of the questions brought about by the
process of the study, from design through fn1ilion .
Although the researcher is satisfied with the interventions, as presented to the

subjects, there are some procedures which may have added to the efficacy of those
interventions. It may be useful to conduct a preliminary study of the coping strategies
used by this group for threat of HIV contraction. A study of coping strategies could
lead to a definition of those patterns of coping that could be detrimenta1 to preventive
behaviour. For example, if one of the strategies is to deny actual risk for contraction,
then it would be necessary to address the denial in prevention programs.
The researcher, s exploration of literature relevant to this study led to a
conclusion that more needs to be done to bring the literature together either by a metaanalysis or by some other means (ie. extensive bibliographic work). There is a vast

87
amount of information regarding AIDS, ranging from epidemiological research to
prevention programs implemented through root systems such as those used in various
Gay and Lesbian Associations as welJ as AIDS !:,>Toups. However, no systematic way
of organizing the literature has heen attempted, as yet. Organizing the research that
has been done in the area of HIV prevention would go far in aiding those who wished
to develop prevention programs or continue rec;earch in this area.
Though the current demographic information provided an in depth exploration
of characteri~cs of the sample groups, more research could he done with this type of
information. For example, gender differences, cultural differences , and/or
environmental differences can be researched in each of the groups with the intention
of determining if either intervention was more effective for a particular gender in
aiding behaviour and attitude change. The open ended questions, designed to look at
types of AIDS knowledge and how that knowledge was acquired, could provide a
valuable beginning to an exploratory study of the levels and paths of knowledge
acquisition of students entering a post secondary institution in northern BC
To determine the efficacy of either of these interventions in changing
behaviour, a longitudinal study should he done following students through several
years. Jn this type of study, several valuable pieces of information could he tested.
Recidivism could be explored, if there is primary change of attitude and/or behaviour,
do the su~jects return to old patterns of behaviour over time.

A recent longitudinal

88
study with interesting ramifications for prevention was done by Caspi et al. ( 1997) and
pointed to the benefit oflongitudinal studies In this study, the researchers found a link
between personality traits and behaviour that was considered risky to health. They
suggested that in the future it may be important to identify personal characteristics of
people in a group and des1gn interventions that address these individual differences .
Due to the fact that HIV 1s not a primary concern for many young people, It may
be more effective to deal with HIV within the parameters of a program geared toward
prevention of all STD's The \Vorld Health Organization ( 1994) reported that a study
done in Zaire demonstrated the eflkacy of STD prevention programs on I IIV
transmission . The types of preventive behaviours necessary to lessen risk for any STD
including HIV is the same. It is possible that a program addressing STD's generally may
be more effective in persuading groups to change behaviour than one directly dealing
with HIV. If the perception of risk for HIV is low, the perception of risk tbr STD
contraction is much higher and. possibly, perceived to be more realistic for the group
addressed in this study Both interventions included in the study could as easily be used
in a general prevention program targeting all STD's.
Conclusion
In the final chapter of this thesis, the study was discussed in general terms. The
fact that Hypothesis I was supported in the area of behaviour was discus<ied and the
relevance of an affective intervention, Intervention I, was explored. Possible reasons tor
the lack of support for Hypothesis 2, in the statistical analyst~. were given and relevance
of an intervention involving members of the target group, in design and implementation

89

of interventions, Intervention 2, was discussed . Perceived limitations of the study were
examined and reasons given as to why these limitations may or may not have had a
significant effect on the outcomes of the study. The final section of this chapter involved
suggestions fbr research that could build on the study and/or research drawn from some
of the perceived limitations of the study
The study had mixed outcomes, according to statistical verification. Interaction
by subjects with a person living with HIV did lead to behaviour change, of a significant
level, and open ended responses showed some of the themes that were strengthened at
the post test level. Involvement of the target group in design and implementation of
intervention(s). did not lead to statistically significant difference in either attitudes or
behaviours Open ended responses did show some change in themes between pre and
post test. However, it would be necessary to do more investigation around this
intervention to ascertain lack of efficacy. The design used here to test the hypothesis
related to this intervention could have been improved upon . The literature does support
the involvement of members of target groups in design and implementation of
interventions and concludes that this involvement is empowering for the individuals and
for the group involved (Bryant et al.. 1992).
Both Intervention I and Intervention 2 should be replicated with more subjects
involved to establish further support for efficacy or lack of eflicacy UNBC presents a
site for further exploration. These interventions could be used within the parameters of a
section on general high risk activities, in such classes as Psvchology and University IO I.
This would allow for testing of efficacy with a larger sample of the population - although

90
not random The interventions could be integrated into a series of psycho-educational
workshops that would fonnulate the framework for an HIV primary prevention program.
These methods could also be used separately during specific times in the calendar year
that are focused on promotion of health and wellness The interventions could be carried
through by members of the Counselling Centre at UNBC, in cooperation with the I Jealth
Nurse on campus who has already done a great deal around HIV prevention, STD's, and
general wellness.
The interventions, that composed this study, could be very useful in fitture design
of prevention programs Even with the mixed results, from analysis of subjects
undergoing the interventions. ~me important information about the way in which
subjects of the target group involved in the study respond to the typei; of designed
strategies was discovered Both interventions have the advantage of replicability in areas
other than HIV prevention or health promotion, therefore, are useful in a variety of
situations, ie. education settings dealing with behaviour difficulties, peer mentoring, and
working to promote more positive behaviour for youth at risk Ultimately, the study
proved to be very valuable and useful.

91
REFERENCES
Abraham, C , & Sheeran, P. (1994) Modelling and modifying young
heterosexuals' HIV-preventive behaviour: A review of theories, findings and educational
implications. Patient Education and Counselling, 24, 199-216.
Ajzen. I. ( 1988) Attitudes, personality and behaviour. Chicago Dorsey Press.
Baffi, C., Schroeder, K., Redican, K., & McCluskey, L. ( I989). Factors
influencing selected heteroxeual male college students' condom use Journal of
American College Heal th, 3 8, 13 7- 14 I.
Baker, S., Morrison, D., Carter, W., & Verdon. M ( 1996) Using the theory of
reasoned action (TRA) to understand the decision to use condoms in an STD climic
population l Iealth Education Quarterly, 23 , "28-542.
Bandura, A. ( 1986) Social foundations of thought and action: A social
cognitive theory. Nev. Jersey Prentice Hall.

Bandura, A. ( 1989) Regulation of cognitive processes through perceived self
efficacy. Developmental Psychology, I 5, 729-735
Bandura. A.. Adams, N.E., & Beyer, J. ( 1977). Cognitive processes mediating
behavior change. Journal of Personality and Social Psychology, 35 125-139.
Baron~R. . & Byrne. D. ( 1987). Social psychology: Understanding human
interaction (Sth ed ). Toronto: Allyn and Bacon.
Beck, K., & Lund, A. ( 1981 ). The effects of health threat seriousness and
personal efficacy upon intention and behavior. Journal of Applied Social Psychology,

il, 401-415

92

Brown, L., DiClementc, R , & Beausoleil, N ( 1992) Comparisons of I Juman
Immunodeficiency Virus related to knowledge, attitudes, intentions, and behaviors
among sexually active and abstinent young adolescents Journal of Adolescent I Iealth,

ll, 140-145
Bryan, A D, Aiken, L.S . & West, S (.1

( 1996).

Jncreasing condom use·

Evaluation of a theory-based intervention to prevent sexually transmitted diseases in
young women. Health Psychology, 15, 371-182
Bryan, A. 0 ., Aiken, L.S ., & West, S G (1997). Young women's condom use
The influence of acceptance of sexuality. control over the sexual encounter, and
percieved susceptibility to common STDs Health Psychology. 16, 468-479.
Burnette, E. ( 1996) AP A, ( DC efforts empower communities to fight AIDS .

APA Monitor [On-line]. Available Internet. http //www apa org/monitor/jun96/
communit.html
Butcher, A , Manning, T., & O' Neal, E. ( 1991 ). HIV-related sexual behaviors
of college students Journal of American College Health. 4, I 15-118.
Canadian Public Health Association. ( 1993 ). Position papers, resolutions and
motions. Health Digest. 17 (Suppl.), 1-12.
Caron, S.L., Davis, C.M., Halteman, W .A., & Stickle, M . (1993). Predictors of
condom-related behaviors among first-year college students. The Journal of Sex
Research, 30, 252-259.
Catania, J.A., Stone, V., Bmson, D & Dolcini, l\il.M. ( 1995). Changes 111
condom use among heterosexuals in wave J of the AMEN survey. J'hc Journal Qf S~

93

Research. 32, 193-200
Centers for Disease Control. ( 1996). General considerations regarding health
education and risk reduction activities In CDC Prevention Guidelines ron-linc)
Available Internet: ... der.cdc.gov/wondcr/ prevguid/p0000389/body0002.
htm#head002002000000000000

Clay, R.A. ( 1996a) Social science increasingly vital in prevention of AIDS.
APA Monitor [On-line]. Available: Internet: http://v.1ww.apa .org/monitor/jun96/

role html
Clay. R A. ( 1996b ). Targeted interventions curb the spread of HIV. AU
Monitor [On-line]. Available: Internet : http://www.apa.org/monitor~jun96/target.htm1
Cohen, D.J., & Bruce, K.E. ( 1997). Sex and mortality Real risk and perceived
vulnerability The Journal of Sex Research, 34, 279-291 .
Cotton, D ., Higgins, D., Person, B., & Darrow, W. ( 1994). CDC behavioral
interventions. American Psychologist. 49, I090-1092 .
Dancy, B. ( 1991 ). The development of an ethnically sensitive and gender-

specific AIDS questionnaire tor African-Amencan women . Health Values. 15, 49-54.
DeBro, S., Campbell, S., & Peplau, L. ( 1994). Influencing a partner to use a
condom: A college student perspective Psychology of Women Quarterly, 18, I6<i- l 82
Eagley, A H., & Chaiken, S. ( 1993 ). Jhe psychology of attitudes. Fort Worth,
TX : Harcourt Brace.

Edgar, T., Fitzpatrick, M.A., & Freimuth~ V.S. (Eds.) ( 1992). AIDS: A

communication perspective. New Jersey: Lawrence Erlbaum Assn

94
Eisen, M., & Zellman, G. ( 1984). Health belief model-based changes in sexual
knowledge, attitudes and behavior. Toronto : ERIC Documents.
Evans, R ( 1989). Albert Bandura: The man and his ideas - Adialogue New
York · Praeger.

Fisher, J., Fisher, W., Misovich, S., Kimble, D , & MalJoy, T. ( 1996) Changing
AIDS risk behavior: Effects of an intervention emphas11:mg AIDS nsk reduction
information. motivation, and behavioral skills m a college student population. Health
Psychology. 15, 1 14-123 .
Fisher, W., Fisher. J , & Rye B. ( 1995) Lnderstanding and promoting AIDSpreventive behavior: 1nsight "i from the theory of reasoned action. Health Psychology,

H, 255-264
Glass, G.. & Hopkins, K. ( 1984) Statistical methods in education and
psychology. Toronto: Allyn and Bacon.
Grube, J., Mayton, D . & Ball-Rokeach, S. ( 1994 ). Inducing change in values,
attitudes, and behaviors· Belief system theory and the method of value selfconfrontation. Journal of Social Issues, 50, 1(iJ-173 .
Hales. D. ( 1994) An invitation to health. (6th ed.). Calitornia:

Benjamin/Cummings Publishing.
Health Canada. ( 1996). HIV in Canada. Surveillance Report. for the Period
l 985-1995. [On-line) Available: Internet: http://hwcwcb.hwc.ca/main/
lcdc/web/publicat/aids/hiv8595/ehiv _a.html
Health Canada. ( 1997a). AIDS and f IIV in Canada HIV/AIDS Epi Update

95

[On-line]. Available· http'//www.hc-sc.gc.ca/hpb/Jcdc/bah/epi/epi I I J _ e htmJ
Health Canada. ( 1997b). Sexual risk behaviors of Canadians. HIV/AIDS Epi

Update [On-line]. Available Internet. http·//www hc-sc gc ca/hpb/ lcdc/bah/epi/
epi6 l l _ e. html# 1

Herlocher, T., Hoff, C, & DeC'arlo, P. ( 1996) Can theory help in HIV
prevention? In HIV: Looking Back, Looking Ahead [On-line]. Available. Internet:
http //chanane. ucsf edu/capsweb/theorytext. html
Holtgrave, D., Doll, L., & Harrison, J. ( I997) Influence of behavioral and
social science on public health policymaking. American Psychologist, 52, 167-173
Howell, D ( 1995). Fundamental statistics for the behavioral sciences (3rd
Ed.). Califon1ia: Duxberry Press.

Iteman1m ( 1993). Mjnnesota: Assessment Systems Corportation.
Jacobs, B., Prentice-Dunn, S , & Reinsch, S. ( 1984). Understanding persistence·
An interface of control theory and self-efficacy theory Basic and Applied Social
Psychology. s, 333-347.
Kanter, F.H ., & Goldstein, A .P. (Eds.) ( 1991 ). Helping people change: A

textbook of methods. New York: Pergamen Press
Kaplan, R., Atkins. C., & Reinsch, S. ( 1984). Specific efficacy expectations

mediate exercise compliance with patients with COPD. Health Psychology, 3, 223-242.
Keeling. R P. ( 199 l ). Time to move forward : An agenda for campus sexual
health promotion in the next decade Journal of American College Health, 40, 51 -53.
Kjng, A., Beazley, R., Warren, W , Hankins, C., Robertson, A., & Radford, J.

96
( 1989) Highlights from the Canada youth and Al DS study Journal of School I Iealth,
59, 139-145

Kowalewski, M R , Henson, K D , & Longshore, D. ( J 997). Rethinking
perceived risk and health behavior: A critical review of HIV prevention research f Iealth
Education & Behavior. 24, J 13-325

Krebs, D , & Schmidt, P. ( Eds ) ( 1993) New directions in attitude
measurement. New York W de Gruyter
KrulJ, C ( J994 ). Level of education, sexual promiscuity, and AIDS. The

Alberta Journal of Educational Research. XL. 7-20.
Learning to~ether about H1Y ( 1994). Alberta . Caring Together Federal
Demonstration Project & The Home Care Program.
Lief, H.J. ( 1989). Human sexuality with respect to AIDS and HIV infection:
Report of a workshop, March 30 to April 2, 1989. Toronto: I Iealth and Welfare

Canada
Mantell, 1 , Divittis, A, & Auerbach, M. ( 1997). Evaluating HIV prevention
interventions New York· Plenum Press
Moskal, R. ( 1991) Effect of a comprehensive AIDS curriculum on knowledge
and atttudinal changes in northern Canadian college students. Journal of Counselling.
~ 338-348

NeedJe, R.H., Brown, BS., Coyle, S.L. , & Weissman, G. (1994) NIDA's HIV
prevention programs American Psychologist. 49, I 089-1090.
O'Leary, A., Goodhart, F., Jemmott, L.S., & Beecher L.D. ( 1992). Predictors

97

of safer sex on the college campus A social cognitive theory analysis. Journal of
American College Health. 40, 254-263.
Orians, C.E., Liebow, E.B, & Branch, K.M. ( 1995) Community based
organizations and HIV prevention among Seattle,s inner-city teens Urban
Anthropology and studies in cultural systems and world economic development. 24, 3658

Ostrow. D , & Kessler, R ( 1993 ). Methodological issues in AIDS behavioral
research. New York: Plenum Press.
PaJys. T. ( 1992). Research decisions. Quantitative and qualitative perspectives.
Toronto: Harcourt Brace.
Petosa, R., & Jackson, K. ( 1991 ). Using the health belief model to predict safer
sex intentions among adolescents. Health Education Quarterly, 18, 463-476.
Pilkington, C, Kern, W, & Indest, D ( 1994). Is safer sex necessary with a
"safe" partner? Condom use and romantic feelings. The Journal of Sex Research, 3 I,
203-210.

Ratzan, S. ( 1993 ). AIDS: Effective health communication for the 90's
Washington: Taylor and Francis.
Rosenstock, I , Strecher, V., & Becker, M. ( 1988) Social learning theory and the
health belief model. I Iealth Education Quarterly, 15, 175- I83.
Ross, M., & McLaws, M. ( I992). Subjective nom1s about condoms are better
predictors of use and intention to use than attitudes Health Education Research, 7, 335339.

98

Rotheram-Borus, M J. ( 1997). Annotation 1ITV prevention challenges-ReaJistic
strategies and early detection programs American Journal of Public Health, 87, 544546

Rothspan, S , & Read, S. ( 1996). Present versus future time perspective and
HIV risk among heterosexual college students. Health Psychology, 15, 13 J-134.
Sandfort, T ., & lessen, G. ( 1992). Denial as a barrier for HIV prevention
within the general population. In Sexual transmission of HJV infection: Risk reduction.
trauma and adaptation, Schonnesson. N (Ed.) New York: The Haworth Press
Sax, G . ( 1989). Principles of educational and psychological measurement and

evaluation (3rd edition). California: Wadsworth.
Seppa, N ( 1997). Young adulls and AIDS 'It can't happen to me.' APA
Monitor [On Line]. Available: Internet : http://www apa org/monitorzjan97/
condo ma. htmJ .

Snider, D.E., & Satcher, D. ( 1997). Behavioural and social sciences at the
Centers for Disease Control and prevention: Critical disciplines for public health.
American Psychologist. 52, 140-142.
Sorenson, J., Wennuth, L. , Gibson, D , Choi, K., Guydish, J., & Batki, S
Prevening AlDS in drug users and their sexual partners. New York : The Guilford Press
Stevens-Smith, P., & Remley, Jr., T., ( I994). Drugs, AIDS, and teeng:
Intervention and the school counselor . .The School Counselor, 4 I, 180-183.

Thurman, Q., & Franklin, K.M ( 1990). AIDS and college health Knowledge,
threat, and prevention at a northeastern university. Journal of American College I lealth.

99
38, 179-183

Torabi, M., & Yarber, W. ( I 992). Alternate forms of I IIV prevention attitude
scales for teenagers AlDS Education and Prevention, 4. 172-182.
Tn1ax, S R ( l 994) On the new call for HIV prevention research Amercian
Psychologjst. 106. 1092-1093 .
Trussler, T ( 1994) \1ore reflections on taking care of each other: health

promotion in community based AIDS work, Vol II. Vancouver: AIDS Vancouver.
Tudiver, F ., Myers, T .. Kunz, R., Orr, K., Rowe, C ., Jackson, E., & Bullock. S
( l 992) The talking sex project: results of a randomized controlled trial of small-group

AIDS education for 612 gay and bisexual men. Evaluation and the Health Professions,

U, 26-42.
Valdiserri, R.O., Aultman, T.V., & Curran, J.W. ( 1995). Community planning:

A national strategy to improve HIV prevention programs. Journal of Community
Health. 20, 87-100.

van der Plight. J., & Richard, R. ( 1994) Changing adolescents' sexual behavior:
percieved risk, self-etlicacy and anticipated regret. Patient Education and Counsellin~.

23, 187-196.
Watney, S. ( 1994). Practices of freedQm: Selected writings on HIV/AIDS.

Durham: Ou ke University Press.
Weinstein, N.D. (1989). Perception of personal susceptibility ofhan11. In V.
Mays, G Albee & S. Schneider (Eds.), Prirnaty prcvention_QCAIDS Psychological

approaches (pp. 142-167).

100

Wiersma, W. ( 1986). Research methods in education. An introduction.

Toronto· Allyn and Bacon.
Wilson, C., & Stewin, L. (1993) The importance ofJevels of knowledge in
developing a unified theory ofHIV/AlDS counsel1ing: A comparison of two culturally
different samples. International Journal for the Advancement of Counselling, 16, 281289

Wolman. B. ( 1989) Dictionary of behavioral science (2nd Ed). Toronto·
Academic Press
Wong-Reiger, D , & Davld, L. ( 1993) A hands-on guide to planning and
evaluation Ottawa· Canadian Hemophilia Society.

World Health Organization (l 994). Report of the strategic meeting on
prevention (HIV and STDs): Geneva. 5-7 October 1994 [On Line] Available. Internet
http ://gpawww.who.ch/summit/prevent. htm
Wulfert, E , & Wan, C K. ( 1993). Condom use. A self-efficacy model. Health
Psychology. 12, 346-353 .
Wulfert, E, & Wan, CK. ( 1995) Safer sex intentions and condom use viewed
from a health belief, reasoned action, and social cognitive perspective The Journal of

Sex Research. 32, 299-3 11 .
Yarber, W (1995) SSSS presidential address: Principles for creating
AIDS/sexuality education messages for youth The Journal of Sex Research, 32, 269274

Zimbardo, P G., & Lieppe, M R ( 199 l ). The psychology of attitude change and

101

social influence. New York: McGraw I lill.

102

APPENDIX A
SAMPLE CONSENT FORM

103

Sample Adult Consent Form for Participation in Experimental Study
Dear Student,
If you are currently a student at UNBC' and are hetween the ages of 18
and 25 I would invite you to take part in an experiment which wiJI occur hetween
September 20, I995 and December 4, 1995. I am a Master'~ student in Educational
Counselling and the re~earch in which I am inviting you to become involved is for my
thesis. My area of interest is in primary prevention of HIV, and the study looks at
two interventions and whether these interventions can be effective in promoting
healthier lifestyles. I am interested in testing methods that may he effective in aiding
people to lessen their risk of contracting HIV.
Your partkipation would involve:
A. Completion of a set of questionnaires before the study.
B. Participation in one of three groups who will each receive a different intervention.
C. Completion of a questionnaire package after the study.
D. Participation in a group discus~ion on the completion of the study.
Your part in this study will require approximately eight to ten hours of your time
spaced over two and a half months.
Anonymity is a must for this experiment, therefore you will find at the top of
this form a number which will be the only identifier of your answer\. If you decide
to take part you will be asked to enter this numhcr as well as the letter of the group to
which you are assigned on each page of the questionnaires. This will allow a match
of the first test with the second test. There will he no way of identifying you with the
numbers you are assigned. During the interventions you will be using your names,
however, the numbers you use for the te~ts will not he mentioned during that time.
If you agree to take part in this study I would like to advise you that it is
essential you not discuss the questions or details of the interventions in which you take
part with other people until the study has been completed. A discussion group will
follow the research to allow you the opportunity to discuss what has occurred, and
your feelings ahout the process.
I would ask you to consider participating in this study and if you have any
4uestions please do not hesitate Lo conta(;t me. Maria Walsh
1

964-2527

Thank you for your time, and .l hope you will consider helping me with this research.

104

APPENDIX B
DEMOGRAPHIC QUESTIONNAIRE

105
Questionnaire # I

Please respond to the following questions Your answers can be placed in the spaces
provided in each question and by circling the correct answer when more than one is
pro\'ided for you There is no right or wrong answer, so please answer as accurately as
possible.

1. I am - - - years old.

2 I am

male

female .

3. The sexual onentation that best describes me is:

3

2

4 I am in my first

totally homosexual

bisexual

totally heterosexual

second

third

4

fourth

5

6

fourth+ year of university.

5 I attended another post-secondary institution before coming to UNBC. yes

no

6 1 entered university the September after I graduated

from secondary school.

7. I attended a secondary school of
A . more than 800 students.

B between 600 and 799 students
C between 400 and 599 students.
D between 200 and J99 students.

E. less than 200 students.

yes

no

106

8. (a) How much information about HlV infection was provided at your secondary
school?
A no information
B very little information
('

some information

D much information

E. complete infonnation

(b) What kinds of information regarding HIV and prevention were you provided
with?

9 (a) J speak with my farru1y members about HIV infection
A. never

B. very rarely
C. occasionally
D. frequently
E all the time

(b) The member of my family J most often talk to about this is

(c) Some of the issues around HIV that we have talked about are:

----------------------

J0. I know person/s who have contracted Hl V

yes

no

107
J l. I speak or have spoken with a partner about safer sex.

yes

no

12 I have been tested for HIV in the past six months

yes

no

yes

no

13 I learned about HIV from

14 My experience with sexual intercourse could best be described as
A. no expenence

B very little experience

C average expenence.
D much experience.

E above average experience.

15. I was ___ years old when I first had sexual intercourse.

16. I speak with my partners about HIV. _

17. I grew up in

a rural area

yes _

an urban area

no

both

18 I speak to friends other than my partners about safer sex.

19. During the semester I have learned this intormation about HIV and safer sex ..

20. (a) I think it is important for me to learn more about HIV .

yes

(b) Why/Why Not?_ _ _ _ _ - - - - - - - _ _ _ _ _ __

no

108

APPENDIX C
A TTlTUDE QUESTIONNAIRE

109

Questionnaire #2

PART A.
This is an opportunjty for you to communicate to me your beliefs about certain areas
of HIV prevention. There are no right or wrong answers. I would like you to
respond to the que!->tions with the response that comes lMMEDIATEJ ,Y to your
mind. Please do not go on to Part B without completing Part A.

1. I believe that the best ways to prevent HJV are ....... .

2. Itjs_/ is not (please circle) necessary for me to practice safer sex practices
because ....

3. The HIV virus affects these groups of people ....

4. What l really believe about HIV and prevention of the virus is ..... .

110

5. What T think about abstinence as a form of prevention is ....

6. These are the things that I feel prevent people from Jiving a preventive lifestyle
(practicing safer sex or abstinance) ...... ...... .

7. This is how my peers feel about preventive lifestyles .............................. ..

8. (a) There are people who do not need to practice safer sex.

(b) If yes, the reason is .....

yes

no

Il I

PARTB.

Please read each statement very carefully and record your IMMEDIATE response by

circling the letter that best represents your opinion There are no right or wrong
answers in this questionnaire, the answer is the one which best represents what you
believe. Please avoid changing answers once you have responded to a statement. The
answers are explained as follows.

VSA - ve1y strongly agree

VSD - very strongly disagree

SA - strongly agree

SD - strongly disagree

A- agree

D - disagree

I . HIV is a punishment for 1mmoral behaviour.

VSA SA A VSD SD D

2. H1V is a virus that is not a concern for the
straight community.

VSA SA A YSD SD D

3. Using a condom during sex can be very erotic VSA SA A

4.

If I were planning to have sex, it would he an
insult if my partner insisted we use a condom. VSA SA

5.

A VSD SD D

Jt is easy to use the prevention methods that
reduce one's chance of getting HIV .

6.

VSD SD D

VSA

SA

A

VSD SD D

I would be 5upportive of a person with
HIV.

VSA SA A VSD SD D

112
7.

I would consider deciding not to have sex as a

form of prevention.

VSA

SA

A

VSD SD

D

The contraction of HIV

VSA

SA A

VSD SD

D

9.

I will use condoms every time I have sex.

VSA

SA

A VSD SD D

I0.

HIV is preventable.

VSA

SA

A

VSD

SD

D

11.

Sharing IV drug needles has
VSA

SA

A

VSD

SD

D

VSA

SA

A

VSD SD

D

8. Monogamy is an important factor in preventing

nothing to do with contraction of HIV.

12. I don't think that speaking to a potential
partner about safer sex is useful .

13. It is not necessary to ask a potential sex partner
to get the HIV antibody test.

VSA SA A VSD SD D

14. No one practices abstinence as a prevention
of HIV infection.

VSA

SA A

VSA

SA

VSD

SD

D

A VSD SD

D

15. People can influence their friends to practice
safer sex behav10urs.

16. The chance of my contracting HIV is so slim

l 13

that if I get it, it must have heen meant to be. VSA SA

A VSD SD D

17. The best way to prevent HIV infection is to
know your partner.

VSA SA

A VSD SD

VSA SA

A VSD SD D

D

18. Women have some difficulties in negotiating
safer sex that men don't experience.
19. If T wanted to I could practice safer sex every
time 1 had sex with someone.

VSA SA A VSD SD D

20. It is not difficult to practice safer sex.

VSA SA A VSD SD D

21. Even if my partner resisted I would still resist
using a condom.

22. AIDS 1s a preventable disease.

VSA SA A VSD SD D

VSA SA A VSD SD D

23 Using a condom during sex decreases
pleasure.

VSA SA A VSD SD D

24. Using someone else's IV needle is an action
that can lead to HIV contraction .

VSA SA A VSD SD

D

25. I would not practice monogamy just to
avoid HIV infection.

26. It is impo1tant to speak to a sex partner about

VSA

SA

A VSD SD D

114
HIV prevention before having sex

VSA

SA

A

VSD

SD

D

VSA

SA

A VSD SD D

VSA

SA

A

VSD

VSA

SA

A

VSD SD

D

VSA

SA

A

VSD

SD

D

VSA

SA

A

VSD SD

D

VSA

SA

A VSD SD D

27. Peer groups are very influentia1 for people
developing attitudes about HIV.

28. I would not be able to practice safer sex in all
of my sexual encounters.

SD D

29. I am not in a high risk group so I don't

have to be concerned about living a
preventive lifestyle

30. Men and women have to negotiate sex in
different ways.

3 1. I would trust potential sex partners who say
they are not infected

32. If I am going to get HIV there is nothing I can

do about it, so why bother.

11 5

APPENDIX D
BEHAVIOUR QUESTIONNAIRE

116

Questionnaire #3

PART A.

Please answer the following questions with as much accuracy as possihle.
Remember there is no right or wrong answer. only answers that reflect your
behaviours. Please finish all of Part A before continuing on to Part B.

1. (a) Do you talk to potential pa,tners about your need to protect yourself
from HIV infection?
yes

no

(b) If yes, how do you do this?

2. In what situations have you had unprotected sex?

3. (a) If you engage in sex what are the safer sex practices you use?

(b) Do you use these safer sex practices all the time?

yes

no

(c) If no, what are the situations in which you do use safer sex and how often do
you use these preventive measures?

4. Have you failed to negotiate for safer sex with your partner but had sex anyway?
Describe the situation including the difficulties you had with negotiation.

5. (a) Have you experjenccd any situation(s) where you have chosen lo abstain from
having sex because you knew you would not be protected? yes no

117
(b) If yes, what helped you decide that this was best for you at the time?

6. (a) Have you taken drugs intravenously?

yes

no

(b) rf yes, how did you protect yourself?

PART B.

Please read each statement careful1y and record your responses by circling
the proper answer. There are no right or wrong answers in this questionnaire, the
answer you give represents YOUR behaviours.
ET = EVERYTIME
ST = SOMETIMES

TO = ONCE OR TWICE
N = NEVER
NA = NOT APPLICABLE TO ME

IN THE PAST SIX MONTHS I HA VE .. .... .
1. Refused to have unprotected sex while I

was rugh or drunk.

ET ST TO N NA

2. Planned and/or used techniques with my partner(s)
that made safer sex more enjoyahle.

ET ST TO N NA

3. Refused to have anal intercourse because there

was no condom.

ET ST TO

N NA

ET

ST TO

N

ET

ST TO

N NA

4. Depended on someone esle to provide me with

their needles so I could shoot up .
5. Had sex wilh only one person

NA

118

6. Talked with friends about HJ V contraction
and prevention.

ST TO

N NA

ET ST TO

N NA

ET

7. Did not practice safer sex becau\e my partner(s)
didn't want to.

8. Asked my partner(s) if they had been tested for HIV . ET ST TO N NA

9. Had unprotected anal intercourse

ET ST TO

N

10. Had sex with a series of partners.

ET ST TO

N NA

11. Asked my partner(s) if they had ever shared needles ET ST

TO

N

NA

NA

12. Depended on my partner(s) to provide protection.

ET ST TO

N NA

13. Had only vaginal intercourse.

ET ST TO

N

NA

14. Shared a needle while shooting up.

ET ST TO

N

NA

15. Had unprotected sex while I was drunk.

ET ST TO N

NA

16. Did not ask my partner( s) if they had the HIV test.

ET ST TO N

NA

ET ST TO

N

NA

18. Tried to discover my partner(s) sexual history.

ET

ST TO

N

NA

19. Had a series of sexual relationships.

ET ST TO

N

NA

20. Chose to remain abstinent.

ET ST TO N

NA

21. Refused to have sex without a com.lorn .

ET ST

TO

N NA

22. Maintained a monogamous relationship.

ET ST TO

N NA

17. Had unprotected sex under pressure from my

partner(s).

119

23. Took responsibility for deciding that a condom
would he used when myself and my partner(s)
had sex.
24. Decided not to engage in sexual intercourse.

ET ST TO N NA
ET ST TO N NA

25. Asked my partner(s) if they had a history of
drug use.

ET ST TO

N

NA

ET ST TO N

NA

26. Told my partner(s) that l wanted to use condoms

when we were having sex.
27. Refused to take part in oral sex in an attempt to

protect myself from HIV.

ET ST TO N NA

28. Refused to engage in anal sex.

ET ST TO N NA

29. Discussed with my partner(s) ways to make

safer sex e~joyahle for us.

ET ST

TO N NA

30. My partner(s) and I discussed our past
sexual histories.

ET ST TO N

NA

ET ST TO N

NA

31 . Had discussed with peers our ideas around
HIV contraction and prevention.