The role of personality, family influences, and prosocial risk-taking
behaviour on substance use in early adolescence
Andrew Wood
B. Psych (Hons)
School of Psychology
Health
Griffith University
Submitted in fulfilment of the requirements of the degree of Doctor of Philosophy in
Clinical Psychology
August 2010
ii
Abstract
This research contributes to contemporary thinking regarding the nature of
impulsivity and its influence on substance use in early adolescence. The current study
examined whether engagement in prosocial risk-taking activities served as mediators
between the more distal factors of personality and family environment, and substance use,
in a large cohort of Grade 8 students (N = 969). This is the first study to examine the
relationship between these variables using a contemporary two-factor conceptualisation
(reward drive and rash impulsiveness) of a personality characterised by impulsivity. The
role of family environment was also directly modelled as this has often been identified as a
key feature in the likelihood of substance use initiation in young people.
Using structural equation modelling (SEM), the results of this research provide
strong support for a two-factor model of impulsivity and the differential influences of rash
impulsiveness and reward drive on prosocial risk-taking behaviours and substance use. As
expected, rash impulsiveness was found to convey direct risk for substance use for males
and females. An interesting finding was that rash impulsiveness did not convey any
influence on participation in physical-risk activities, however had a strong negative
association with performance-risk activities and prosocial behaviour more broadly.
Although reward drive did not convey direct risk for substance use, this facet of impulsivity
was found to be strongly and directly associated with engagement in physical- and
performance-risk activities as well as prosocial behaviour more broadly. Participation in
physical-risk activities was associated with greater substance use. These results suggest that
reward drive has had an indirect influence on substance use through participation in
physical risk-activities. Finally, turning to the role of family influences, family environment
iii
did not convey any significant direct influence on substance use, however, may have an
indirect influence through the association with rash impulsiveness and reward drive. The
data suggest that a positive family environment was associated with greater prosocial
behaviour, but not with participation in physical- and performance-risk activities. No
significant gender differences in the hypothesised model were found. For males and
females, the hypothesised model accounted for 16% (CI95: .09 - .21) and 21% (CI95: .12 .28) of the variance in substance use, respectively.
iv
Declaration
This work has not previously been submitted for a degree or diploma in any
university. To the best of my knowledge and belief, the thesis contains no material
previously published or written by another person except where due reference is made in
the thesis itself. The data for this research was collected and analysed by the author.
_______________________
Andrew P. Wood
Date: 30 August 2010
v
CONTENTS
Abstract ....................................................................................................................... ii
Declaration................................................................................................................. iv
List of Figures ............................................................................................................. x
List of Tables ............................................................................................................. xi
Acknowledgements ................................................................................................. xiii
CHAPTER 1 ........................................................................................................................... 1
Overview and Aims of the Thesis .............................................................................. 1
CHAPTER 2 ........................................................................................................................... 6
Conceptualising the Problem ...................................................................................... 6
Prevalence of substance use in young people ................................................. 7
Australian secondary school students’ use of alcohol .................................... 7
Australian secondary students use of illicit drugs ........................................ 14
CHAPTER 3 ......................................................................................................................... 17
Factors Involved in Early Substance Use and Abuse ............................................... 17
Age of substance use initiation ..................................................................... 18
vi
Concurrent behavioural problems as a risk factor ........................................ 19
Family risk factors ........................................................................................ 24
Personality as a key risk factor ..................................................................... 30
Peer and school risk factors .......................................................................... 41
Social engagement and prosocial activities as protective factors ................. 42
CHAPTER 4 ......................................................................................................................... 54
Definitional and methodological issues in the literature on adolescence and
substance use. ........................................................................................................... 54
Measuring adolescent substance use ............................................................ 54
Contemporary thinking regarding the construct of impulsivity ................... 60
Revisiting the role of impulsivity in substance use: Rash impulsiveness &
reward drive .................................................................................................. 69
Summary....................................................................................................... 73
CHAPTER 5 ......................................................................................................................... 75
Aims and Hypotheses of the current study ............................................................... 75
Method ...................................................................................................................... 77
Participants ................................................................................................... 77
vii
Procedure ...................................................................................................... 77
Measures ....................................................................................................... 79
Substance Use ............................................................................................... 79
Impulsivity .................................................................................................... 83
Behavioural and Emotional Functioning ...................................................... 86
Prosocial Risk Taking................................................................................... 88
Family Functioning ...................................................................................... 88
Results ...................................................................................................................... 91
Data Analysis ................................................................................................ 91
Data Screening.............................................................................................. 93
Reliability and Validity ................................................................................ 95
Descriptive Information ................................................................................ 99
SEM Analysis : Testing the Hypothesised Model ...................................... 106
Multi-group Analysis : Combined Groups ................................................. 113
Model modifications ................................................................................... 126
CHAPTER 6 ....................................................................................................................... 130
viii
Discussion............................................................................................................... 130
Prevalence of substance use ....................................................................... 131
The influence of personality on substance use ........................................... 133
The relationship between personality, physical- and performance-risk
activities and prosocial behaviour .............................................................. 137
The influence of family environment on substance use and prosocial
behaviours ................................................................................................... 141
The relationship between physical- and performance-risk activities,
prosocial behaviours and substance use ..................................................... 143
Strengths of the current research ............................................................................ 146
Age of participants...................................................................................... 146
Model fit ..................................................................................................... 146
Limitations of the current research ......................................................................... 147
Priorities for future research ....................................................................... 149
Summary and concluding comments .......................................................... 151
REFERENCES ................................................................................................................... 153
APPENDIX A .................................................................................................................... 183
ix
APPENDIX B ..................................................................................................................... 187
APPENDIX C ..................................................................................................................... 188
APPENDIX D .................................................................................................................... 190
APPENDIX E ..................................................................................................................... 192
APPENDIX F ..................................................................................................................... 194
APPENDIX G .................................................................................................................... 211
APPENDIX H .................................................................................................................... 212
x
List of Figures
Figure 3. 1. Structural model of the effects of parental processes on antisocial peer
association and substance initiation.......................................................................... 25
Figure 3. 2. Pathways to drug use identified by Giancola & Parker (2001) ......................... 34
Figure 5. 1. Hypothesised model and structural paths. ......................................................... 76
Figure 5.2. Model 2 Males with standardised path coefficients. ........................................ 111
Figure 5.3. Model 2 Females with standardised path coefficients.. ................................... 117
Figure 5.4. Model 2 Combined (males) with standardised path coefficients.. ................... 119
Figure 5.5. Model 2 Combined (females) with standardised path coefficients.. ................ 120
Figure 5.6. Model 2 Combined with Modifications (males) with standardised path
coefficients.............................................................................................................. 128
Figure 5.7. Model 2 Combined with Modifications (females) with standardised path
coefficients.............................................................................................................. 129
xi
List of Tables
Table 2. 1. Proportion of students using alcohol in their lifetime, previous month, or
previous week. ............................................................................................................ 8
Table 2. 2. Secondary school students’ average number of alcoholic drinks per week ......... 9
Table 2. 3. Recent alcohol use in young people ................................................................... 10
Table 2. 4. Most common sources of alcohol and average number of drinks, and
relationship between most common places for drinking and average number of
drinks ........................................................................................................................ 12
Table 2. 5. Australian secondary school students’ lifetime and past month use of illicit
substances ................................................................................................................. 13
Table 3. 1 Measures used in 6-year prospective study of adolescent males (Giancola &
Parker, 2001) ............................................................................................................ 33
Table 4. 1. Summary of factor analytic studies from Dawe and Loxton (2004). ................. 68
Table 4. 2. Summary of studies using measures loading onto an RI-like factor .................. 70
Table 5. 1 Sample Demographics ......................................................................................... 78
Table 5. 2. Summary of means, standard deviations and Cronbach’s alpha coefficients of
key Measures. ........................................................................................................... 96
Table 5. 3. Factor weights for the substance use measures .................................................. 97
Table 5. 4. Factor loadings for the Prosocial Activities Scale (PAS) ................................... 98
Table 5. 5. Self-reported patterns of alcohol use ................................................................ 100
Table 5. 6. Self-reported patterns of tobacco use ............................................................... 101
Table 5. 7. Self-reported lifetime cannabis use .................................................................. 101
xii
Table 5. 8. Means, standard deviations and group differences between males and females
on measure of substance use. .................................................................................. 102
Table 5. 9. Means, standard deviations and group differences between males’ and females’
participation in prosocial risk-taking activities. ..................................................... 102
Table 5. 10. Males reported frequency of participation in physical- and performance-risk
activities .................................................................................................................. 103
Table 5. 11. Females reported frequency of participation in physical- and performance-risk
activities .................................................................................................................. 104
Table 5. 12. Means, standard deviations and group differences between males and females
on individual and family measures. ........................................................................ 105
Table 5. 13. Bivariate correlations, means, and standard deviations for males.................. 107
Table 5. 14. Model fit indices for the male group .............................................................. 109
Table 5. 15. Non-significant paths from Model 1 Males.................................................... 109
Table 5. 16. Non-significant paths from Model 2 Males.................................................... 110
Table 5. 17. Bivariate correlations, means, and standard deviations for females .............. 114
Table 5. 18. Model fit indices for female group ................................................................ 115
Table 5. 19. Non-significant paths from Model 1 Females ................................................ 115
Table 5. 20. Non-significant paths from Model 2 Females ................................................ 116
Table 5. 21. Bivariate correlations, means, and standard deviations for the total sample.. 118
Table 5. 22. Model fit indices for combined groups. ......................................................... 122
Table 5. 23. Non-significant hypothesised paths for Model 2 Combined (females). ......... 122
Table 5. 24. Non-significant hypothesised paths for Model 2 Combined (males). ............ 123
Table 5. 25. Model fit indices for combined groups. ......................................................... 127
xiii
Acknowledgements
I wish to express my sincere and heartfelt appreciation to my supervisor, Professor
Sharon Dawe, for supporting me through the development of this thesis and assisting me to
establish the study upon which it is based. It has been a privilege to have been surrounded
by Sharon’s clinical and research knowledge and expertise for so many years. This, along
with her warmth and kindness has ensured an otherwise daunting task was made possible.
I also wish to extend my gratitude to Dr Matthew Gullo for his advice, assistance
and patient support with data analysis.
I would not have been able to complete this without the support of my wife Nicole,
who has endured every step of this journey with me. She has spent countless weekends
parenting our children alone while I was studying. I thank her with all my heart for her
love, support and encouragement.
I would also like to thank the many other people who have provided support and
assistance along the way; my associate supervisor Dr Stefano Occhipinti for his statistical
advice and guidance; Dr Analise O’Donovan for her kind words and gems of wisdom over
the years; Dr Sally Frye for her motivational interviewing skills; Dr Paul Harnett for his
advice during the early stages of my research; Julie Nos for her assistance with data
collection and for continuing the project; Bernadette and Brian Wood, my parents, for their
encouragement and support, and helping us with the kids; Max and Dawn McKenzie, for
supporting Nicole and I and helping us in so many ways; all the students who have
participated in this study, the school principals who supported it, and the teachers who so
professionally organised the data collection sessions.
To Elijah, Emiliana, and Sasha. You can have your dad back now!