AN ANALYSIS OF THE RELATIONSHIP BETWEEN
SCHOOL BONDING AND DRUG USE AMONG HIGH SCHOOL
STUDENTS

by
Annette Vogt
Bachelor of General Studies
EXTENDED ESSAY PROJECT SUBMITTED
IN PARTIAL FULFILLMENT OF
THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF ARTS IN CRIMINAL JUSTICE

In the
School of Criminology and Criminal Justice

Annette Vogt 2009
UNIVERSITY OF THE FRASER VALLEY
Winter 2009

©All rights reserved. This work may not be
reproduced in whole or in part, by photocopy
or other means, without permission of the author.

Approval
Name:

Annette Ruth Vogt

Degree:

Master of Arts (Criminal Justice)

Title of Thesis:

An analysis of the relationship between school bonding
and drug use among high school students

Examining Committee:
Chair:

Name
[Correct title – Consult your Grad Secretary/Assistant

___________________________________________
Name
Senior Supervisor
Correct title – Consult your Grad Secretary/Assistant
___________________________________________
Name
Supervisor
Correct title – Consult your Grad Secretary/Assistant
___________________________________________
Name
[Internal or External] Examiner
Correct title – Consult your Grad Secretary/Assistant
University or Company (if other than SFU)
Date Defended/Approved:

___________________________________________

1

ABSTRACT
This study examined the relationship between social bonding indicators and drug
use among grade eight to twelve students, drawing from attachment and involvement
dimensions of Hirschi’s Social Bonding Theory for the conceptual framework. Data was
utilized from the Safe School Social Responsibility Survey for Secondary Students
(SSSRS) for the 2005/06 and 2006/07 school years to understand the extent to which
school bonding, a measure of attachment to school and involvement at school, was
associated to current and future drug use. ‘Attachment to school’ indicators included
perceptions of students’ liking for school, feeling respected from adults at school, and
feeling their ideas were important to adults at school. ‘Involvement at school’ indicators
included self-reported rates of students’ participation in groups or clubs at school, and
non-participating behaviours involving skipping class, and skipping an entire day of
school. This study provides important evidence, as well as practical indicators, to support
the proposition that school bonding is associated to, and predictive of, current and future
drug use.

2

ACKNOWLEDGEMENTS
I wish to thank the following people who have provided on-going encouragement
and support to help make the completion of this study possible.
I am deeply appreciative to have had the opportunity of completing this study
under the supervision and mentorship of Dr. Darryl Plecas. His unwavering passion and
enthusiasm for research and educational advancement was not only infectious, but
allowed me to enjoy the challenges along the way. My own appreciation for the practical
application of such a knowledge-building exploration was amplified by Dr. Plecas’ strong
desire to make research relevant at the community level. Thank you for this richness.
I also wish to thank Dr. Terry Waterhouse for encouraging me to make the most
of this study. His background knowledge of the educational literature prompted me to
develop a more thorough articulation of the key topic areas. I am also grateful to Dr.
Waterhouse for granting me access to the survey data through his leading position as the
faculty researcher for the Safe School Social Responsibility Survey project.
I appreciate Dr. Irwin Cohen’s willingness to be available for some of the
methodological discussions. The generosity of his time spent to offer expertise in this
important area was instrumental in developing and addressing the way in which the study
progressed.
I also wish to acknowledge the dedication and commitment of the many school
district staff and students who facilitated and provided valuable information that allows
studies such as this to move forward.
Finally, my family and friends have provided a steady source of background
support in numerous ways. My mother, Adelaide, tirelessly edited papers and sustained
me during peak times by providing support on the home front. My father, George, and
sons, Brad and Anson, have shown an interest in my work and ‘cut me some slack’
during my studies. My close friends have encouraged me to play along the way,
providing me with an avenue of reprieve from the intensity of working and studying.

3

TABLE OF CONTENTS
Abstract...............................................................................................................................2
Acknowledgements ............................................................................................................3
Table of Contents ...............................................................................................................4
List of Figures and Tables.................................................................................................6
Chapter 1: Introduction ....................................................................................................8
Health and Educational Harms Associated to Drug Use..................................................8
Problematic Drug Use ......................................................................................................9
Adolescence ...................................................................................................................11
School Connectedness....................................................................................................12
Chapter 2: Literature Review.........................................................................................15
Risk Factors for Substance Use......................................................................................18
Individual Risk Factors...............................................................................................18
Family Risk Factors....................................................................................................21
Peer Risk Factors........................................................................................................22
Neighborhood and Societal Risk Factors ...................................................................26
School Risk Factors....................................................................................................27
Chapter 3: Theoretical Framework of School Bonding ...............................................32
Social Control Theory ....................................................................................................32
Hirschi’s Social Bonding Theory...................................................................................32
Social Bonding Theory and Schools ..............................................................................35
Risk and Protective Factors for School Bonding ...........................................................36
Individual Risk and Protective Factors.......................................................................37
Family Risk and Protective Factors............................................................................38
Peer Risk and Protective Factors ................................................................................39
School Risk and Protective Factors............................................................................40
Chapter 4: Methodology..................................................................................................46
Theoretical Orientation ..................................................................................................46
The Safe School Social Responsibility Survey for Secondary Students........................47
Data Collection Process .................................................................................................48
Dependant Variable: Drug Use ......................................................................................49
Independent Variable: School Bonding .........................................................................50
School Attachment .....................................................................................................50
School Involvement....................................................................................................52
Data Analysis .................................................................................................................53
Specifics about Sample Characteristics..........................................................................56

4

Chapter 5: Key Findings .................................................................................................60
Associations for School Bonding (Attachment Measures) and Drug Use for
Time 1 (2005/06 – current year)....................................................................61
School Attachment Dimensions .................................................................................61
School Involvement Dimensions................................................................................67
Associations for School Bonding in Time 1 (2005/06) and Subsequent Drug
Use in Time 2 (2006/07)................................................................................72
School Attachment Dimensions .................................................................................72
School Involvement Dimensions................................................................................79
Chapter 6: Discussion and Conclusion ..........................................................................85
Objective 1 .....................................................................................................................85
School Attachment .....................................................................................................86
Involvement................................................................................................................87
Objective 2 .....................................................................................................................88
School Attachment .....................................................................................................88
Involvement................................................................................................................89
Implications for Schools.................................................................................................90
Limitations .....................................................................................................................95
Conclusion......................................................................................................................96
Appendix A – Safe School Social Responsibility Survey for Secondary
Students...........................................................................................................................112

5

LIST OF FIGURES AND TABLES
Figure 1: Gender Comparison of Matched Sample to Unmatched Sample.......................57
Figure 2: Grade Comparison of Matched Sample to Unmatched Sample.........................57
Figure 3: Ethnic Comparison of Matched Sample to Unmatched Sample ........................58
Figure 4: Educational Aspiration Comparison between Matched Sample to
Unmatched Sample ...........................................................................................58
Table 1: Differences between students who reported on whether or not they like
school and drug use at Time 1 (2005/06)..........................................................61
Table 2: Differences between students who reported on whether or not adults at
their school respected them and drug use at Time 1 (2005/06). .......................63
Table 3: Attachment: Differences between students who reported on whether or
not they felt their ideas and opinions were important to at least one
adult in their school and drug use at Time 1 (2005/06). ...................................65
Table 4: Differences between students who reported on whether or not they
skipped a class and drug use at Time 1 (2005/06). ...........................................67
Table 5: Differences between students who reported on whether or not they
skipped all day and drug use at Time 1 (2005/06). ...........................................69
Table 6: Differences between students who reported on whether or not they
participated in a school club or group and drug use at Time 1
(2005/06). ..........................................................................................................70
Table 7: Differences between students who reported on whether or not they liked
school at Time (2005/06) and drug use at school at Time 2 (2006/07). ...........73
Table 8: Differences between students who reported on whether or not they liked
school at Time (2005/06) and drug use in the community at Time 2
(2006/07). ..........................................................................................................74
Table 9: Differences between students who reported on whether or not they felt
adults in their school respected them at Time 1 (2005/06) and drug use
at school at Time 2 (2006/07). ..........................................................................75
Table 10: Differences between students who reported on whether or not they felt
adults in their school respected them at Time 1 (2005/06) and drug use
in the community at Time 2 (2006/07). .............................................................76
Table 11: Differences between students who reported on whether or not they felt
their ideas and opinions were important to at least one adult in their
school at Time 1 (2005/06) and drug use at school at Time 2 (2006/07). ........77

6

Table 12: Differences between students who reported on whether or not they felt
their ideas and opinions were important to at least one adult in their
school at Time 1 (2005/06) and drug use in the community at Time 2
(2006/07). ..........................................................................................................78
Table 13: Differences between students who reported on whether or not they
skipped a class at Time 1 (2005/06) and drug use at school at Time 2
(2006/07). ..........................................................................................................79
Table 14: Differences between students who reported on whether or not they
skipped a class at Time 1 (2005/06) and drug use in the community at
Time 2 (2006/07)...............................................................................................80
Table 15: Differences between students who reported on whether or not they
skipped all day at Time 1 (2005/06) and drug use at school at Time 2
(2006/07). ..........................................................................................................81
Table 16: Differences between students who reported on whether or not they
skipped all day at Time 1 (2005/06) and drug use in the community at
Time 2 (2006/07)...............................................................................................82
Table 17: Differences between students who reported on whether or not they
participated in a school club or group at Time 1 (2005/06) and drug use
in school at Time 2 (2006/07). ..........................................................................83
Table 18: Differences between non-drug using students who report on whether or
not they participated in a school club or group at Time 1 (2005/06) and
drug use in the community at Time 2 (2006/07). ..............................................84

7

CHAPTER 1: INTRODUCTION
Drug use among youth can lead to serious consequences affecting health and
educational outcomes. It leads to increased costs in healthcare from hospitalizations,
shortens lifespan, and contributes to more than 21% of all deaths when considering the
general Canadian population fifteen years and older (Rehm, Ballunas, Brochu, Fischer et
al. 2006; cited in Canadian Centre on Substance Abuse, 2007b). Though substance use
is associated to a variety of problems regardless of age, adolescents are at particular risk
since “young people tend to use substances more often and in riskier ways than older
people” (CCSA, 2007c). Canadian youth, as compared to youth from other countries,
appear to be at even greater risk for harm due to heightened exposure to substance use
(Adlaf & Racine, 2005; Currie, Roberts, Morgan et al., 2004). Figures indicate that 60%
of all illicit drug users are between fifteen to twenty four years of age (CCSA, 2007a). In
response to the seriousness of this problem, the Canadian Centre on Substance Abuse
(CCSA) is currently leading a national strategy to reduce the prevalence of illicit drug use
among youth, with a focus on addressing risk and protective factors associated to drug
use (CCSA, 2007a).

Health and Educational Harms Associated to Drug Use
The harms associated with licit and illicit drug use are numerous and may result in
damage to self, others, and property. Certain consequences are immediate, while others
have long-term impacts on physical, psychological, and social health that may not be

8

initially evident (Gottfredson & Hirschi, 1990; Robertson, David & Rao, 2003; United
Nations, 2005). Oral or injection use is linked to over-doses and the spread of viruses,
and in particular, alcohol ingestion is linked to accident rates from drunk driving, falling,
or other similar outcomes (Health Canada, 2001). The long-term use of alcohol
contributes to liver disease, cancer, heart disease, brain damage, and dependency, while
chronic use of alcohol impedes healthy brain development, cognitive processing,
memory, and poor health, including weight gain (Health Canada, 2001). Smoking results
in a greater likelihood of developing respiratory problems and disease, cancer, and heart
disease (Health Canada, 2001).
High rates of drug use among youth have become a central area of concern for
educators as well (Zyngier, 2003). Substance use leads to problems in school, including
poorer learning outcomes, lower academic aspirations, and premature school leaving
(Hall, Doran, Degenhardt & Shepard, 2006). School counsellors and psychologists have
identified substance use being linked to major mental health problems among youth in
schools (Romer & McIntosh, 2005; in LaRusso, Romer & Selman, 2007), and a small but
noteworthy proportion of criminal careers (eg. vandalism, theft, burglary) begin during
the school years from activities associated with drug use (Akers & Lee, 1999; Roberts et
al. 2001).

Problematic Drug Use
The Diagnostic and Statistical Manual of Mental Disorders (DSM-IV) (2000)
distinguishes substance abuse from use. Substance abuse goes beyond experimentation or
recreational use to become problematic when a maladaptive pattern results in impairment
or distress over the course of a year. Maladaptive behaviour exists when it interferes with
9

a major obligation, such as school (eg. poor performance, suspensions, or expulsions); is
physically dangerous to do so (eg. driving a car; recurrent involvement with the legal
system for substance-related behaviour); and, leads to recurring personal, interpersonal,
or other social problems (Centre for Addiction and Mental Health, 2007). Of youth who
report substance use, only a portion (3-16%) meet the criteria for DSM diagnoses of
substance use disorders, varying by age, gender, and type of substance used. The highest
rates occur for older youth, among males, and with those who have marijuana and
alcohol-related disorders (Weinberg, Rahdert, Colliver et al. 1998; cited in Weiner,
Abraham & Lyons, 2001).
From a legal perspective, a further distinction concerning drug use is made
between illicit and licit drugs. Illicit drugs refer to banned substances as identified in
international drug control agreements (Hall, Doran, Degenhardt & Shepherd, 2006). This
includes, for example, cannabis (e.g. marijuana, hashish), stimulants (e.g. cocaine,
methamphetamine), dance-party drugs (e.g. ecstasy, MDMA), opiates (e.g. heroin,
opium), and pharmaceutical opiates (e.g. methadone, morphine). Licit drugs, by
comparison, are those that are legally obtainable, and include, for example, alcohol and
cigarettes. However, their legal purchase and distribution is accompanied by regulations
that restrict their use by youth. Despite regulations in Canada, survey research involving
Canadian youth fifteen to nineteen years of age identified that twenty percent of
adolescent drinkers had reported at least one harm resulting from an incident of a
physical, social or legal nature. This compared to 30% of youth who reported similar
forms of injury from illicit drug use (ter Bogt, Schmid, Gabhainn, Fotious et al., 2006; in
CCSA, 2007b).

10

Not only is the use of drugs illegal by minors, but the dangers associated to
substance use experimentation in childhood and adolescence can result in immediate and
long term impairment. Early experimentation with alcohol has shown to be a robust
predictor of alcohol abuse and dependence later in life. Figures indicate that 40% of
adolescents who started drinking at fourteen years of age or less had eventual alcohol
dependence, whereas only 10% did so when reporting that their drinking began after age
twenty (Dawson, 1997; in Health Canada, 2001). While experimentation and asserting
independence is part of normal human development (Canadian Health Network, 2007),
risk-taking through exposure to, and early initiation of, alcohol and drug use can
compromise healthy growth and development (Health Canada, 2001).

Adolescence
Adolescence is a developmental period characterized by rapid changes in the
physical, psychological, and social facets of a young persons’ life (CCSA, 2007b;
Simons-Morton, Crump, Haynie & Saylor, 1999). This period of life sets the stage on
which life-long skills and behavioural patterns are established. In addition to the
considerable tension and financial burden for families affected by their child’s
involvement with drugs and alcohol (CCSA, 2007b), substance use contributes to
developmental problems. It impedes healthy transitions by increasing chances that
children will leave home early, drop out of school, or face parenting responsibilities at a
young age (Krohn, Lizotte & Perez, 1997). School counsellors and psychologists have
identified that substance use is linked to major mental health problems (Romer &
McIntosh, 2005; in LaRusso, Romer & Selman, 2007). A small but noteworthy
proportion of criminal careers begin during the school years from increased exposure to
11

negative peer influences associated to drug involvement (e.g. vandalism, theft, burglary)
(Akers & Lee, 1999; Health Canada, 2001). The younger the age of onset of drug use, the
greater the chance for developing serious problems over time (Hawkins, Arthur &
Catalano, 1995; Jarvinen & Gundelach, 2007; McCrystal, Percy & Higgins, 2007;
Mosher, Rotolo, Phillips, Krupski & Stark, 2004). There is a need to prevent youth from
spiralling into positions of disadvantage from the harmful effects of early and on-going
contact with drugs and alcohol (CCSA, 2007a).
Some youth, however, believe drug use is ‘beneficial’. They maintain that it
enhances mood, offers relief from pain, provides an outlet for asserting independence,
allows them to engage in a pleasurable or novel act, satisfies curiosity, and helps them
gain entry into a social group (CCSA, 2007b; Department of Social and Economic
Affairs, 2005; Health Canada, 2001; Hunt, Evans & Kares, 2007; Robertson, David &
Rao, 2003). It is important to acknowledge the underlying conditions and motivations
that lead to experimentation with drugs in order to understand how society can best
support healthy child and adolescent development (Health Canada, 2001).

School Connectedness
Research has identified that students’ connectedness to school is one of the
greatest protective factors against polydrug use, absenteeism, pregnancy, and
unintentional injury, even after taking family influences into account (Bonnie, Britto,
Klostermann et al. 2000). School connectedness has also been identified as a foundational
element linked to the successful transfer of pro-social attitudes, norms, and values among
youth (Larson, 2000). However, scholars have also noted a form of toxicity in schools
referred to as “a contagious youth culture of academic negativism and misconduct”
12

(Simons-Morton et al. 1999:99). Harsh school discipline practices, lack of adequate
support from school staff, failing to set and maintain clear behavioural boundaries, and
ignoring individual differences have been linked to undesirable behaviour among youth
(Mayer, 1995; in Lewis, Sugai & Colvin, 1998). One of the most significant determinants
affecting behaviour lies in the quality of interactions that students develop with school
staff (Resnick, Bearman, Blum et al. 1997, cited in Bonny, Britto, Klostermann, Hornung
& Slap, 2000:1017), suggesting that schools play an important role in shaping behaviour.
This study will draw from Hirschi’s Social Bonding Theory to establish an
analytic framework for considering the role of schools in mitigating risk by examining
the situations and type of changes that schools can make in preventing early and on-going
use of drugs. Hirschi stated that deviance and eventual criminality would occur when an
individual’s bonds to society were weak or broken. Along with family and peer networks,
he specifically included students’ relationship to school in his application of the theory.
This study introduced the construct of school bonding by targeting two of Hirschi’s four
dimensions of social bonding theory: attachment and involvement. This was done to
determine whether selected measures school attachment and school involvement were
associated to, and predictive of, current and future drug use among high school youth.
To answer this question, the current study examined student self report data from
an existing data set involving grade eight to grade twelve students in two medium-sized
British Columbia school districts over two consecutive years (2005/06 and 2006/07).
Specifically, the study sought to; 1) confirm the extent to which school bonding was
associated to drug use behaviour, as well as; 2) examine the extent to which school
bonding was a preceding condition to drug use behaviour. The findings are intended to

13

assist in gaining a better understanding for the role of schools in addressing risk and
protective factors related to drug use among students for the development of more
effective prevention-based policies and practices.

14

CHAPTER 2: LITERATURE REVIEW
International rates of substance use among youth peaked in the 1980’s, declined
in the early 1990’s (Johnson, O’Malley, Bachman & Schulenberg, 2006; Hawkins,
Arthur & Catalano, 1995), and subsequently rose again (United Nations, 2005). The
Monitoring the Futures study (Johnson, O’Malley, Bachman & Schulenberg, 2006)
measures substance use among American youth, reporting that more than half (55%) of
young people in 1975 used illicit drugs by the time they had left school. By 1981, rates
rose to 66%, followed by a decline to 41% in 1992. Since the drop in the 1990’s, drug use
has risen again, fluctuating between 55% in 1999 to current day rates of 48% in 2006.
The 2005 World Youth Report documents this current rise, drawing specific attention to
the use of synthetic drugs that is occurring on a worldwide scale (United Nations, 2005).
Canadian substance use patterns are similar to international patterns (Health
Canada, 2001). A Health Canada report, Preventing Substance Use Problems Among
Young People: A Compendium of Best Practices, identifies substance use patterns for
Canadian youth by comparing findings from a number of studies conducted in various
provinces between 1994 to 1999. It provides an important reference document for this
study. One of these studies involves the Ontario Student Survey on Drug Use (OSDUS),
being the longest continuous Canadian survey of youth substance use. It shows that
substance use peaked in 1979, followed by a steady decline until the early 1990’s, at
which time the use of licit and illicit drugs followed a consistent upward trend (Adlaf,
Paglia & Ivis, 1999; cited in Health Canada, 2001). Apart from alcohol and tobacco, the

15

drug use rates mirrored those from the 1970’s when substance use had peaked in Canada
(Brounstein & Sweig, 999; in Health Canada, 2001).
On average, more students used drugs than did not in Canada (King, 1999; in
Health Canada, 2001). This general finding was identified from self-report data gathered
from grade seven to grade thirteen high school students during the 1990’s through the
International Health Behaviours of School-Age Children study (1999). Cross-sectional
data from the 1998/1999 National Longitudinal Survey of Children and Youth (NLSCY),
using results from twelve to fifteen year old students found average rates of alcohol use
were 42%; with 22% having drunk to intoxication; 19% having used cannabis; and, 11%
having tried hallucinogens at least once in the previous year (Hotton & Haans, 2004).
Since these figures only apply to students from the public education system, the figures
will not necessarily reflect prevalence rates among youth who were home schooled, from
private schools, had skipped or dropped out of school, or were incarcerated (Health
Canada, 2001). Among some of those groups outside the public education system,
prevalence rates may be even higher.
Patterns of drug use among Canadian youth vary by type of drug. In their review
of various drug use literature, Health Canada (2001) reported alcohol to be the most
commonly used substance, with approximately two out of three youth reporting use. This
was followed by tobacco and cannabis, which were reported at the rate of approximately
one in three. Students from British Columbia had the highest use of cannabis, with 40%
of students having tried it at least once in the previous year. West Coast youth were also
noted as the highest national users of drugs in 2007 by Vancouver Coastal Health having
found that among those between the ages of sixteen and twenty five years of age, 90%

16

reported alcohol use, 70% marijuana use, and more than 50% had tried cigarettes
(Vancouver Coastal Health, 2006).
The Health Canada report states that hallucinogens were the fourth most common
substance of choice among Canadian youth. Depending on province, hallucinogens were
used at the rate of between 9% and 14%, and LSD at the rate of 6% to 11%. Far less
common, were the use of inhalants, stimulants, cocaine, methamphetamine, heroin, PCP
and non-medical use of other drugs, including tranquillizers and barbiturates. The latter
group of drugs ranged in use from 3% to 10%, depending on the specific province, or
other group characteristics (Health Canada, 2001).
A comparison study (Hotton & Hans, 2004) of substance use rates between
Canadian and American youth examined results from the 2002 Youth Smoking Survey
(Canadian) and the 2002 Monitoring the Futures Study (American). Findings indicated
that more Canadian than American youth used alcohol (57% and 47% respectively), with
cannabis use being similar (17.1% and 19.2% respectively). A striking difference,
however, was found with the use of inhalants among grade eight students: 7% of
Canadian youth reported its’ use as compared to 15% of American youth.
Inhalant use is a growing problem among American youth, and is linked to first
experiences with a psycho active drug (Mosher et al. 2004). This is believed to be due, in
part, to the availability of inhalants through a wide range of household items (nail polish,
paint thinner, and aerosols) that allow youth to experience a ‘cheap high’. Native
American youth had the highest level of use. While inhalant use was most common in
early teens (National Institute of Drug Abuse, 1998; cited in Mosher et al. 2004),
continued use declined with age. It is suggested that the apparent decline may, however,

17

be masked by mortality rates inherent through the limitations of using cross-sectional
school-based survey data (Mosher et al. 2004).
In Canada, the use of “club drugs”, including ecstasy, methamphetamine,
Rohypnol, GHB, and Ketamine, though less frequent, is reported by the Ontario Student
Survey on Drug Use (OSDUS) at 4.4% across the age range of students (Adlaf, Paglia &
Ivis, 1999; cited in Health Canada, 2001). This reflected a between-grade usage ranging
from approximately 1% of seventh grade students to 10% of eleventh grade students who
reported use over the past year. The increase has been of particular concern due to the
significant rise from 1993, when only 1% claimed to have used club drugs across grades.

Risk Factors for Substance Use
Risk factors contributing to the use of drugs and alcohol among youth are
documented at the individual, family, peer, school and community level.

Individual Risk Factors
Gottfredson and Hirschi (1990) state that delinquency arises from a lack of self
control. They argue that delinquency and substance use are inter-related behaviours that
predict criminogenic tendencies. Ones’ lack of self control stems from a desire for
interim gratification through avenues that require less effort, and result in more
immediate and concrete rewards (Hwang & Akers, 2003). This suggests that substance
use is a matter of individual desire, choice or free will.
Others challenge the notion of free will by acknowledging the influence of
physical and chemical impairments that may cause setbacks and create additional
challenges for individuals (Health Canada, 2001). This alternate view suggests that some

18

individuals face certain barriers that add a layer of complexity regarding drug use
(CCSA, 2007b). Biological interference during brain development, for example, may
result from family involvement in drug use (Health Canada, 2001). Biological
interference may also lead to psychological impairments becoming evident through the
development of mental disorders, challenging temperaments, child abuse, or trauma
(Glantz & Pickins, 1992; in CCSA, 2007b). Individuals who are challenged by such
factors, as found in many street youth or those on their way to becoming street youth, are
more likely to become early users, heavy users, and drug abusers (CCSA, 2007b; Health
Canada, 2001; Robertson, David & Rao, 2003). Rather than simply viewing delinquency
as a behavioural choice stemming from a lack of self control, chemical or physical
impairments may contribute to the use of drugs and alcohol as a way of coping with a
series of problems that may confront certain youth (Health Canada, 2001).
Another view suggests that substance use is linked to social disadvantage. One
form of social disadvantage associated to problematic drug use stems from disrupted
childhoods, particularly for those who have a history of school failure and subsequent
involvement with crime (Hawkins, Arthur & Catalano, 1995; Buchanan, 2005; in
McCrystal, Percy & Higgins, 2007). Heavy and multi-drug use is much more likely to
occur among sexually-abused and exploited youth, youth from socially disadvantaged
situations, refugee youth (United Nations, 2005), gay/ lesbian and questioning teens,
youth in custody, runaway and street involved youth, adolescents with co-occurring
disorders, and First Nation, Inuit and Metis youth (CCSA, 2007b). Cumulative levels of
disadvantage at the individual level may fuel further marginalization and subsequent
patterns of drug use through a process known as ‘othering’, where individuals or groups

19

of individuals (including certain neighborhoods) may become viewed as being somehow
different (and therefore alienated) from the rest of society (McCrystal, Percy & Higgins,
2007).
Research links substance use patterns with age. Substance use peaks in the early
twenties and then declines (Gottfredson & Hirschi, 1990). Tracing this shift further, the
2005 National Survey on Drug Use and Health (NSDUH) reported usage rates between
the ages of twelve to eighteen years of age jumping five-fold for drugs, and thirteen-fold
for alcohol within this six year developmental period (U.S. Department of Health and
Human Services, 2005). In Canada, the 1998/99 National Longitudinal Survey of
Children and Youth (NLSCY), based on 4,296 respondents aged twelve to fifteen years,
asked students to report their level of drug and alcohol use “at some time”. Regarding
alcohol, 17% of twelve year olds admitted to its use, compared with 66% for fifteen year
olds. Drinking to intoxication increased from 4% for twelve year olds to 29% for fourteen
year olds, and 44% for fifteen year olds. Marijuana use grew from 3% for twelve year
olds to 38% among fifteen year olds (Hotton & Haans, 2004). The average age of first
initiation for alcohol for a significant minority of young people in Canada was around
fourteen years of age (CCSA, 2007a), with patterns reflecting first use of marijuana in
grades eight to grade nine, and other substances in following years (Health Canada,
2001).
Competing views (Tittle, 1995; cited in Akers & Lee, 1999) exist concerning the
extent to which age is a useful concept for explaining, predicting, or preventing substance
use. Substance use must be viewed more fully by considering the broader forces,
including the social and developmental factors associated to age. Combining Tittle’s

20

(1995; Tittle & Ward 1993, cited in Akers & Lee, 1999) premise that the causes of crime
and delinquency are the same regardless of age, social learning theory differentiates the
effects of age that interact differently according to ones’ place in the social structure.
Accordingly, age has an influence on normative patterns of behaviour driven by direct
and indirect social interactions, imitating others, reinforcements, and surrounding
attitudes about certain behaviours (Akers & Lee, 1999). The peer, family, church, or
school contexts tend to have a direct influence on behaviour but to differing degrees that
depend on the developmental stage of life (Akers & Lee, 1999). Thus, it appears as
though age may be too simplistic a notion when seeking to understand and address this
form of risk-taking in adolescence.
Individual factors leading to the use or abuse of drugs is a complex issue. Young
people use drugs for a variety of reasons. While many youth use drugs for their perceived
benefits or availability, marginalized youth may be more driven to use drugs to escape
their current situation, memories of the past, or to fulfil the need to escape negative
emotions, including depression or suicide ideation (CCSA, 2007b; Health Canada, 2001).

Family Risk Factors
Youth from disadvantaged homes that are distinguished by parental conflict, poor
supervision (Hotton & Haans, 2004), hostile or negative parenting styles, lack of
structure, inconsistent application of rules, or where anger and threats occur with great
regularity, are more likely to use alcohol or drugs (Hall, Doran, Degenhardt & Shepard,
2006). Youth from step-parent families, compared to intact families, were found to be
nearly two times more likely to use drugs (Hotton & Haans, 2004). Additionally, youth
who come from homes characterized by lack of attachment to parents or caregivers, and
21

where caregivers abuse substances, are at increased risk for using drugs and alcohol
(Robertson, David & Rao, 2003). Home life, however, becomes less influential as
children grow toward young adulthood; at which point they become vulnerable to other
factors. This may include risks associated to (un)employment, or relationships with an
intimate partner (Hoeve, Smeenk, Loeber et al. 2007). It appears as though the strength of
family influences associated to adolescent substance use varies with development.

Peer Risk Factors
Youth may be drawn to use drugs and alcohol due to attitudinal influences from
peers, their tendency to imitate peers, and through reinforcement from peers (Hotton &
Haans, 2004). Results from the 1999/1998 National Longitudinal Survey of Children and
Youth (NLSCY) show youth to be eleven times more likely to use alcohol when
reporting that most of their peers consumed alcohol, and nearly twice as high for drinking
to intoxication among youth who claimed to have friends who were often in trouble. The
same general pattern held true for the use of drugs (Hotton & Haans, 2004). In another
study, (Skinjner & Wilborn; cited in Faircloth & Hamm, 2005) individuals with peers
who conformed to conventional norms of behaviour and who remained committed to
academic success had higher rates of school achievement. On the otherhand, studies also
show that having peers who were not engaged in school, nor motivated by its’ processes
and goals, were much more likely to report delinquent behaviour, including
experimentation with drugs and alcohol use, gang affiliation, and violence. Though the
directions of these associations are unclear, peer connections appear to play a reinforcing
and influencing role affecting decision-making about the use of drugs and alcohol
(Garcie-Reid, Reid & Peterson, 2005; Goodenow, 1992; Hotton & Haans).
22

Identity formation for youth, including their need to belong to a peer network, is
one of the reasons why youth decide to use drugs. International research (Jarvinen &
Gundelach, 2007) examining alcohol-related lifestyles gathered information from two
hundred Danish grades nine to grade twelve students. They wanted to know why youth
defined certain behaviours, such as using drugs and alcohol, as being valuable or right
(Jarvinen & Gundelach, 2007). Youth gave a high level of symbolic capital to the use of
alcohol through participating in the party culture. They also gave a high level of social
distinction to those who engaged in alcohol use to the point of intoxication because it
allowed them to be noticed, and even admired for their behaviour among certain groups.
In turn, those who participated in this form of lifestyle were subsequently invited to other
parties, while those who did not embrace a drinking lifestyle stayed home. The authors of
this study noted an ‘alcohol-centric’ dichotomous and conditional pathway toward
belonging that provided continued justification for the on-going use of alcohol and for
some, drugs as well. Not surprisingly, the study was also able to determine that those who
became high-risk drinkers were also those who were more likely to experience more
emotional distress and who had poorer relationships with their parents. This finding
confirms earlier research reported by Health Canada (2001) suggesting certain youth
were at increased risk for substance abuse as a way of dealing with stress, though it does
not identify causality.
The apparent growing normalization of drugs and alcohol in society has
contributed to a social climate among peers that may encourage use. Past research
comparing attitudes about the use of drugs stems back to the 1990’s when tolerance for
substance use grew (Hawkins, Arthur & Catalano, 1995). Adlaf, Paglia and Ivis’s

23

research (2001, cited in Hotton & Haans, 2004), using the OSDUS, reported a 19%
increase in acceptance for the use of marijuana between 1991 and 2001 (42% vs 61%
respectively). The report showed a similar pattern for the “once or twice” use of cocaine,
rising in acceptance from 55% of the population from its previous rate of 41%. As views
became more tolerant, corresponding drug use rates, including heavy drinking, also rose.
Youth Court Statistics also mirrors the rise in drug possession and trafficking cases for
the same period (Hotton & Haans, 2004). Similar patterns have been noted across Canada
(Hotton & Haans, 2004), and more globally (Substance Abuse and Mental Health
Services Administration, 2002). For example, a Belfast study (McCrystal Percy &
Higgins) uncovered very recent trends to suggest that “drug use is no longer restricted to
“delinquency or street corner ‘no hopers’”(2007:48), but that a broader proportion of
youth have “assimilated and legitimated recreational drug use” (2007:48). Some suggest
the rave or party scene is a cultural marker representing a shift from its former
association with the subculture (Parker et al. 1998b; cited in McCrystal, Percy & Higgins,
2007), to becoming more in-grained into the broader culture of society (Blackman, 2004;
cited in McCrystal, Percy & Higgins, 2007).
The social climate concerning drug use among peers is linked to the growing
normalization of drugs, and supported by a risk management logic concerning its use.
The CCSA (2007b) has acknowledged the widespread use of drugs, attributing age, and
the extent to which youth acknowledge the potential harm of drugs associated to use.
After examining the attitudes and behaviours of three hundred middle-class, well
educated, rave-attending youth in the San Francisco Bay area, Hunt, Evans and Kares
determined that substance use behaviour is “far from being passive, ignorant or ill-

24

informed of the dangers… [but that young people] operate and actively engage in an
elaborate system of techniques designed to minimize a sense of uncertainty associated to
their drug use “(2007:77). For example, while fear of having a bad trip was identified as a
central concern, youth explained a decision-making rationale that involved an assessment
about the level of social trust among their friends, their own frame of mind, drugassociated risks, and the overall context in which the activity would occur.
Youth rank ordered drugs by level of dangerousness (Hunt, Evans & Kares,
2007). The greatest perceived threat to well-being was methamphetamine, heroin, and
cocaine. Ecstasy was seen to be low risk. The following statement captured a commonly
held rationale for the use of ecstasy:
I think that it’s one of the lesser harmful drugs out of like the drug spectrum.
As long as you make sure you don’t get dehydrated, or get bad stuff, you’re
gonna be all right…You’re not gonna overdose off of it, as long as you don’t
take a whole bunch of it. And…you know, it’s usually a pretty good
experience (Hunt, Evans & Kares, 2007:84).
Though a sense of risk was admittedly present, drug use appeared to be of no greater
consequence than environmental pollutants that youth were exposed to on a daily basis.
One youth remarked, “I have concerns about...the long term effects of me breathing
smog...so I don’t...concern myself with...the use of a drug more than any other thing in
my environment” (Hunt, Evans & Kares, 2007:85). Most youth agreed that drug use was
generally worthwhile (Hunt, Evans & Kares, 2007). While certain perceived benefits of
substance use may appear to outweigh the risks for youth who use drugs experimentally
or even occasionally, research states that a substantial sub group of young people will not
only harm their own well-being but place others at risk through drug-related activities
(CCSA, 2007b).

25

Neighborhood and Societal Risk Factors
The increased use of drugs and alcohol among youth mirrors a societal shift in
norms found among adults who model greater tolerance toward alcohol and drugs (Health
Canada, 2001; McCrystal, Percy & Higgins, 2007). While societal attitudes reflect a
broader backdrop effecting decision-making about drugs, deprived and socially excluded
communities have been linked to some of the most problematic drug use (Eith, 2004).
Neighborhood cohesion is the extent to which a community can articulate and agree to
common values of its people, and then mobilize to solve its’ own problems (Sampson &
Groves, 1989; in Kohen, Dahinten, Leventhal & McIntosh, 2008). However, as
neighborhood cohesion decreases, drug use increases. Disadvantaged communities are
more likely to lack an ability to monitor their children and youth (Sampson, 1997, in
Kohen et al. 2008), increasing the likelikhood of increasing adolescent delinquency due
to unsupervised peer group activities (Sampson & Groves, 1989; in Kohen et al. 2008),
poor parenting practices, the lack of positive role models, and the absence of positive
socializing influences from institutional resources (eg. libraries, parks) that provide
formal and informal relationship building opportunities between individuals, families,
and the community (Kohen et al. 2008). The quality of interactions and level of cohesion
between youth, adults, and the broader community is critical for effective monitoring
children during two major transition periods: as children progress from elementary to
secondary; and, when young adults leave home to attend college or go to work
(Robertson, David & Rao, 2003).

26

School Risk Factors
Schools represent a primary developmental arena where students’ concept of self,
peer connections, interpersonal skills, habits and worldviews are formed. A primary
purpose for establishing a public education system was to inculcate a cohesive set of
norms and values for the regulation and construction of human capital, and to keep
children out of trouble (Wotherspoon, 2004). The public education system was founded
on middle class values and norms (Eith, 2004) that include “ambition, individual
responsibility, self-denial, rationality, delaying of gratification, industry, manners, control
of aggression, wholesome recreation, and respect for property” (Gottfredson & Hirschi,
1990).
Given their primary position as socializing structures for youth, social justice
advocates maintain that conventional educational models fail to account for the extra
needs of students who face greater challenges attaining the normative standards for
behaviour and prized academic performance goals (Akers & Lee, 1999; Eith, 2004). This
suggests that students are under varying degrees of tension and stress to conform to the
educational and behavioural standards in society (Hoppe, Wells, Haggerty, Simpson, et
al. 1998).
Critical theorists argue that vulnerability to substance use is linked to the creation
of an underclass in schools through exclusionary and marginalizing practices (Furlong &
Cartmel, 1997; cited in McCrystal, Percy & Higgins, 2007). Exclusionary practices have
been linked to labour market forces that drive the need for higher educational attainment.
Students who are cut off from opportunities to develop successful career trajectories by
the ages of sixteen and eighteen are more likely to find themselves living in an

27

“undercaste” lifestyle and unable to participate in a largely consumer driven society
(Furlong & Cartmel, 1997; in McCrystal et al. 2007:36).
Schools may unwittingly contribute to a risk environment as they operate within a
paradigm based on equality of opportunity rather than through equity-based frameworks
(Collier, 2006; Eith, 2004; Mortimore, 1999). Equality of opportunity fails to recognize
that all students do not all begin school, nor proceed through life, from a similar place of
‘advantage’. Students from socially and economically disadvantaged positions face
greater challenges in meeting or complying with established educational standards for
behaviour, academic performance, and even outward appearance (Akers & Lee, 1999).
Though behavioural science has long wrestled with finding the most efficient method for
educating society’s youth, Nel Noddings (1992) contends that an over-riding focus on
efficiency has been guided by an ideology of control. This has been dominated by a
predominant emphasis on curricular instruction as the primary goal of schooling that
results in alienating students at the cost of developing caring relationships between the
individual teacher and the individual student to buffer children from risk (Noddings,
1992).
Automation and resulting student alienation can be observed through the practice
of streaming. Streaming is a commonly used tiered educational process based on ‘ability’
groupings (Eith, 2004). It results in the sorting of students into either an academic stream
(headed for university preparation) or a non-academic stream (characterized by technical
or vocational training) (Eith, 2004).Critics claim that streaming segregates and alienates
students based on their performance, rather than working with students’ potential (Eith,
2004; Wotherspoon, 2004). Alienation may occur as the effects of streaming detract from

28

students’ sense of well-being through a competitive-driven system that generates
‘winners’ and ‘losers’. When students perceive schooling to be unfair, their growing
disillusionment may result in disengagement from school and create a climate of risk for
engaging in drug use (Eith, 2004; Gottfredson & Hirschi, 1990; Mortimore, 1999, cited in
Kane, 2006).
Scholars acknowledge the limited impact that school effectiveness research has
made since Willis’ (1977; cited in Kane, 2006) seminal ethnographic work, Learning to
Labour: Why working-class kids get working-class jobs was published. At that time,
Willis concluded that the failure of schools to make advances in gaining compliance,
particularly by boys from working-class identities, is because ‘working-class’ students
have often made a rational decision to reject ‘compliance” for ‘credentials’. This is the
case for males from lower class backgrounds who are more likely to group with similar
circumstanced peers to assert their status through an inverted value system (Eith, 2004).
Research suggests the system needs to “permit as many as possible to succeed – albeit at
different speeds, with different amounts of support and to different levels” (Mortimore,
1999; cited in Kane, 2006:675).
The best behavioural outcomes occur as students personally identify with the
norms and goals of their school (Rutter , Maughan & Mortimore et al. 1979, cited in
Henry & Slater, 2007). While attention and awareness about codes of conduct are
important, much of what is seen to be problematic behaviour (such as drug use) are more
accurately linked to high rates of boredom, alienation and disconnection from a
meaningful sense of challenge at school (Larson, 2000). In addition to engaging students
through strengthening relational ties, schools play an important role to promote social

29

cohesion by establishing and maintaining firm sanctions regarding substance use. Like
parenting, schools can provide safe environments by engaging youth at their level,
challenging them, and through careful monitoring of behaviours (Voeklk & Frone, 2001).
As risk factors associated to substance use among youth accumulate across
individual, peer, neighborhood, and school domains, young people are confronted with
compounding levels of risk (CCSA, 2007b; Hawkins, Arthur & Catalano, 1995). This has
important implications for designing prevention and intervention strategies to reduce
early and on-going drug use. Though young people may perceive or realize certain
benefits associated to substance use, prevention strategies to reduce and delay early onset,
as well as to reduce and prevent the use of drugs and alcohol, must take into account the
potential for risks at the individual, family, peer, school and community level.
The continued proliferation of drug education programs in schools today suggests
that additional exploration is needed to understand the on-going problem of adolescent
drug use. Various approaches since the 1960’s have been tried. These range from among
harm reduction, reducing the supply of substances through regulation and law
enforcement, reducing demand for supply, and offering school and community-based
prevention programming (Hawkins, Arthur & Catalano, 1995). Though the scientific
evidence for substance use prevention education programming is inconclusive (Cohen,
Plecas & Watkinson, 2005; Gandhi, Murphy-Graham, Petrosino, Chrismer & Weiss,
2007; Health Canada, 2001), there is growing agreement about the importance of
adopting a risk-focussed perspective that draws attention to the importance of
strengthening protective factors in young peoples’ lives (Hawkins, Arthur & Catalano,
1995; Scales, 2005; Schaps & Solomon, 2003).

30

The research focussing on the reduction of risk through emphasizing protective
factors is an emerging approach for prevention science (Hawkins, Arthur & Catalano,
1995; Health Canada, 2001). While social problems tend to arise from the compounding
effect of risk-related influences, protective factors have the opposite effect (Hawkins,
Arthur & Catalano, 1995; Schaps & Solomon, 2003). They may lessen risk, provide a
buffer against risk, interfere with risk, or prevent dysfunction from occurring (Coie, Watt
& West et al.1993; in Hawkins, Arthur & Catalano, 1995). Rarely do substance abuse
prevention programs target the social context of an adolescent’s school environment
adequately (Henry & Slater, 2007; Schaps & Solomon, 2003). Since many of the risk
factors surrounding youth are difficult to influence directly, schools are important sites
for investigating whether protective factors can be strengthened through more focused
attention toward the individual student, the individual teacher, and their relationship in
order to reduce substance use involvement.

31

CHAPTER 3: THEORETICAL FRAMEWORK OF SCHOOL
BONDING
Social Control Theory
While also recognizing the value that overlapping theories, such as learning
theory, choice theory, and structural theory may contribute to an examination of
delinquent behaviour, the current study will utilize social control theory. Social control
theory is interested in knowing why people obey the rules of society, while assuming that
all people have the potential and opportunity to commit crime or delinquency.
Accordingly, though all people may be enticed to commit crime, some are better able to
resist enticement than others (Siegal & McCormick, 1999). Travis Hirschi’s Social
Bonding Theory (1969) was one of the first expressions of social control theory (Siegal &
McCormick (1999).

Hirschi’s Social Bonding Theory
Travis Hirschi (1969) stated that deviance and criminality arises when an
individual’s bonds to society are weak or broken. An individuals’ motivation to act in a
pro-social manner is motivated by fear of damaging important relationships with societal
others. Social Bonding Theory is also one of the earlier theories to explore delinquency in
the context of an individuals’ relationship with broader societal structures, such as
schools (Eith, 2004). The reference to ‘societal’ entities is broad, ranging from
individuals (parents, teachers, parents, peers) to groups (families, schools,
churches)(Hirschi, 1969). Thus, without the development of a social bond, and without an
32

internal desire for social approval from important external others, people are free to
commit criminal or deviant acts (Hirschi, 1969).
The ‘social bond’ includes four parts: attachment, involvement, commitment, and
belief. Attachment is viewed as a central psychological mechanism that connects an
individual to a significant societal ‘other’ (Hirschi, 1969). Attachment involves having an
interest in, and sensitivity to, the opinions of others (Hirschi, 1969). For the same reason
that attachment promotes a moral conscience through the internalization of norms, the
psychopath is free to deviate. The psychopath lacks moral restraint, feeling no guilt or
fear of disapproval by societal others when he behaves in anti-social ways because he
lacks a conscience or super ego (Hirschi, 1969).
The second element, involvement, refers to time spent doing activities of a
conventional nature (Hirschi, 1969). The degree to which one becomes involved in
activities reduces opportunities to engage in delinquent behaviour, crowding out the
possibility for even conceiving of such acts (Hirschi, 1969). Involvement has a
reinforcing and insulating effect that reduces delinquency because youth who are
engaged in conventional activities tend to be under the supervision or guardianship of
supportive adults (Hirschi, 1969).
Belief is the extent to which an individual embraces the established norms and
values to be morally legitimate and personally desirable (Hirschi, 1969). Views and
opinions that favour sensitivity toward others, and showing respect for the legal code
constitute examples of the belief component (Siegel & McCormick, 1999). Belief
influences behaviour depending on how firmly one belongs to a social network of people

33

who share a common understanding of, and live according to, those conventions (Siegel
& McCormick, 1999).
Commitment refers to the time, energy and effort that an individual invests in
certain types of activity (Hirschi, 1969). While wanting to attain a university education is
an example of a prized conventional goal, commitment involves the choices and effort
required to attain it. As individuals are faced with an opportunity to engage in deviant
behaviour, they must also consider the cost that deviant action may place on jeopardizing
that goal (Hirschi, 1969). The commitment element reinforces social bonding by
providing a form of holding power (or resiliency) toward a certain line of activity (Eith,
2004).
Assessments of Social Bonding Theory generally support the notion that weak
societal bonds increase the likelihood of delinquency (Hawkins, Arthur & Catalano,
1995). Opposition to Hirschi’s theory has given greater clarity to the way in which the
various elements function. For example, higher levels of involvement do not necessarily
result in less delinquency. What appears to matter is that the object of attachment is prosocial in nature (Burton, Cullen, Evans et al. 1995; cited in Siegel & McCormick, 1999,
Hirschi, 2001; in Hirschi, 1969). A similar finding was true for attachment. Hirschi’s
earliest supposition stating that any form of social attachment was favourable, regardless
of whether attachments were formed with deviant entities, has been discredited
(Hindeland, 1973, Jensen & Brownfield, 1981, Samuelson, Hartnagel & Krahn, 1990,
Warr, 1993; cited in Siegel & McCormick, 1999).
Further testing has shown social control variables to be less predictive of male
behaviours than females, suggesting the basis for this lies in the theory’s orientation

34

toward social relations (Krohn & Massey, 1980, Rosenbaum & Lasley, 1990; cited in
Siegel & McCormick, 1999). Influencing factors appear to differ by developmental age.
A study examining a group of grade six to eight students in New Brunswick found selfesteem and physical health to be the most significant individual level factors influencing
school belongingness (Ma, 2003). In another study, weak bonds to parents and teachers
during mid-teen years (15 years), and strong bonds among males at twelve and eighteen
years of age, led to delinquency (LaGrange, Pandina & LaGrange, 2006). Finally, though
Hirschi claimed the causal chain toward criminality originated from weak bonds, Robert
Agnew (1985) identified that the direction of travel could occur in reverse order as well
(cited in Siegel & McCormick, 1999). Despite its criticisms, Social Bonding Theory has
been an enduring framework for understanding early forms of delinquency and crime
(Siegel & McCormick, 1999).

Social Bonding Theory and Schools
In the context of schools, scientific testing generally supports the notion that when
students feel personally connected to school, they are less likely to engage in destructive
behaviours, more likely to internalize social norms, and better able to succeed in school
(DeWit, Akst, Braun et al. 2002; Voelkl & Frone, 2001). Furthermore, the cumulative
effect of social bonding found schools with higher collective levels of bonding to school
reporting fewer instances of delinquency among students, including their use of drugs
and alcohol (Henry & Slater, 2007).
Diaz (2005) identified school attachment as a feeling of being a part of a school
through meaningful connections. School connectedness is referred to as “an adolescent’s
experience of caring at school and sense of closeness to school personnel and the
35

environment” (Resnick, Bearman, Blum et al. 1997, cited in Bonny, Britto, Klostermann,
Hornung & Slap, 2000:1017). School attachment is “the extent to which students ‘feel’
that they are embedded in, and a part of, their school communities” (Johnson, Crosnoe &
Elder, 2001; cited in Diaz, 2005:320). Similarly, school attachment is “the extent to
which [students] feel personally accepted, respected, included and supported by others in
the school social environment (Goodenow, 1992:4). Common to all definitions is a
reference to the interconnected nature between the individual (the student) and the social
environment at school. All definitions point to a resulting emotive state arising from
students’ experiences and perceptions of their social environment. Taken together, school
attachment culminates in the individuals’ sense of connectedness to school personnel
(Diaz, 2005), school processes (Johnson, Crosnoe & Elder, 2001, as cited in Diaz, 2005),
and how they perceive being treated (respected, included, and supported) by others at
school (Goodenow, 1992). This reinforces the notion that students’ level of attachment to
school stems from both individual and broader school level factors linked to the social
climate of their school (Maddox & Prinz, 2003; Cernkovich & Giordano, 1992; cited in
Eith, 2004).

Risk and Protective Factors for School Bonding
School bonding is complex. The literature draws attention to an interaction of life
domains at the individual, peer, and school context that expose youth to multiple levels of
risk for developing healthy bonds to school.

36

Individual Risk and Protective Factors
Students who report a lower sense of belonging to school have higher overall
rates of substance use. DeWitt et al. (2002) found that about two-thirds of students report
a strong sense of belonging to their school, while 15% strongly disagreed. The
consequences of students’ lack of belonging has been associated with loss of motivation,
academic success, general delinquency, premature leaving from school, and substance
use (Enomoto, 1997; Garcia-Reid, Reid & Peterson, 2005; Marsiglia, Miles & Dustman,
2003). A study examining a population of Native youth found that sense of belonging
played a protective role against drug use even when accounting for differences in school
achievement (Napoli, Marsiglia & Kulis, 2003).
Substance use inhibits school engagement (Garcia-Reid, Reid & Peterson, 2005)
and leads to cognitive impairment (McCrystal, Percy & Higgins, 2007). If caught under
the influence at school, students risk being expelled (McCrystal, Percy & Higgins, 2007).
Academic competence, on the other hand, is cited as a protective factor because some
students view drug use as interfering with academic success ( David & Rao, 2003).
Results from the 1998/1999 National Longitudinal Survey of Children and Youth found
youth who reported poor grades in school, in comparison to those who reported good
grades, were more than twice as likely to drink to intoxication. Similarly, those indicating
a weak commitment to school were more inclined to report having been intoxicated
(Hotton & Hans, 2004). Interpersonal experiences with school personnel are critical for
promoting healthy development, pro-social behaviour, and school success (Akers & Lee,
1999; Wentzel, 1989; in Faircloth & Hamm, 2005).

37

Family Risk and Protective Factors
The extent to which children and youth feel connected, committed, and successful
in school is mediated by family involvement (Eith, 2004). Parents or adult caregivers
contribute to school belonging and academic success through consistent efforts in
monitoring their children, showing interest in homework, and becoming involved in the
day-to-day activities of the school (Garcie-Reid, Reid & Peterson, 2005).
Specific to boys, a moderate inverse association between low support and
neglectful parenting style and school achievement was identified (Sampson & Laub,
1993, 2001; cited in Hoeve, Smeenk, Leber et. al. 2007). In addition, males are more apt
to develop beliefs that favour delinquency (Zhang, Loeber, & Stouthamer-Loeber,1997;
in Pardini, Loeber & Stouthamer-Loeber, 2005). This phenomenon strengthens in
adolescence (Larson & Richards, 1991; cited in Pardini, Loeber & Stouthamer-Loeber,
2005) and is further exacerbated among youth who have a relationship with caregiver(s)
who adopt negative communication patterns and harsh discipline methods (Kochanska,
1997; in Pardini, Loeber & Stouthamer-Loeber, 2005). Relational processes that depend
on compliance-based motivation, rather than fostering the internalization of pro-social
values through positive interactions, may predispose boys, particularly in the middle
school years, to drift from family in search for new peers who support their delinquent
actions (Carlo et al., 1999; in Pardini, Loeber & Stouthamer-Loeber, 2005). Furthermore,
while family ties were found to strengthen pro-social behaviour during early childhood
and adolescence, social connections through work or a marriage relationship during
young adulthood had a greater influence on behaviour (Sampson, 2001; in Hoeve,
Smeenk, Leber et. al. 2007). This suggests that family connectedness may play a
decreasing role with age (Sampson & Laub, 1993; in Hoeve, Smeenk, Leber et. al. 2007).
38

This creates an opportunity for schools to strengthen their connections with youth
through interventions that facilitate positive peer connections, expose youth to adult role
models, and assist parents in their ability and effectiveness to monitor the social
relationships of their children (Pardini, Loeber & Stouthamer-Loeber, 2005).
School attachment is associated with opportunity. It is more likely for students
from positions of poverty and other forms of disadvantage to lack a sense of connection
to their school (Bornholt, 2000; cited in Napoli, Marsiglia & Kulis, 2003; Collier, 2006;
Skinjner & Wilborn, 1994; cited in Faircloth & Hamm, 2005). Children and youth from
economically disadvantaged family systems have greater difficulty acquiring social,
technical and academic skills that are foundational to school success (Eith, 2004).
Poverty increases pressure for adult caregivers to meet daily requirements for survival,
leaving few resources for furthering the intellectual and social development of their
children. While already being faced with learning difficulties and lower academic
performance (Collier, 2006), students from positions of disadvantage face further risk
through experiencing higher rates of social alienation at school (Collier, 2006).

Peer Risk and Protective Factors
School attachment is influenced through peer networks. Students who are not
engaged in school, nor motivated by institutional processes and goals, are more likely to
seek belonging with less conforming peers involved in drug and alcohol experimentation,
gang affiliation, or violence (Garcia-Reid, Reid & Peterson, 2005). On the other hand,
peers who demonstrate a commitment to academic success and who conform to the
dominant norms facilitate academic achievement in others (Skinjner & Wilborn, 1994; in
Faircloth & Hamm, 2005). A successful preventative approach for steering a young
39

Latino student away from violence and destructive behaviour was to stimulate stronger
connections between the youth and their family, school, and community (Peacock,
McClure & Agars, 2003; cited in Diaz, 2005). While peer connections are an important
factor influencing behaviour, pressure from friends generally failed to over-ride the
influence that supportive and accepting teachers could contribute within the school
setting (Garcie-Reid, Reid & Peterson, 2005; Goodenow, 1992). Promoting school
attachment through enhancing social support from teachers for a group of high-risk
students appears to reduce incidents of self-destructive, risky, or antisocial behaviour.

School Risk and Protective Factors
Teacher Care
“... the single greatest complaint of students ..is... they don’t care!”
(Comer, 1988; cited in Noddings, 1992:2).
The importance of caring teachers must not be under estimated. Caring teachers
foster resiliency in children through a relationship of mutual respect (Anderman, 2003;
DeWit et al. 2002). A founding educational philosopher, Nel Noddings, articulated an
ethic of caring in l982 for the profession of teaching, tracing her concept of care to a
German philosopher, Martin Heideggar (1962). He stated that care was “the very Being
of human life” (1962; cited in Noddings, 1984:15) that required a state of consciousness
that allowed the ‘carer’ to become “seized by the needs of the other” (1962; cited in
Noddings, 1984:16). This form of care, however, is criticized for the way it suggests that
assistance is handed down from one person to another, rather than a form of mutually
beneficial exchange of energy between the ‘cared’ and the ‘cared for’ (Noddings, 1984).
Noddings (1984) acknowledges the tension that can exist for teachers in balancing

40

control of the classroom and demonstrating care for students. Teachers fulfil an important
role helping students access their potential by adopting pedagogies that “engage and
enable students in valued and worthwhile activities, linking learning not just to the
community, but also empowering students to use their own authentic knowledge, values
and culture” (Gale, 2000; in Zyngier, 2003: 136). Healthier forms of demonstrating care
in schools allows teachers to impart confidence in their students about their rightful place
in society, particularly among those faced with extraordinary disadvantage (Collier,
2006). Noddings maintains that to feed children’s spirits is a higher order of
professionalism brought about by encouraging mutual exchange, building relationships,
and fostering sense of belonging.
Through her work with Latino students, Valenzuela (1999) expands on Noddings
notion of care by demarcating aesthetic care from authentic care (Antrop-Gonzalez and
De Jesus, 2006). Aesthetic care requires students to surrender to the teachers’ preference
for “things and ideas”, regardless of cultural or social chasms that may exist between
them. This form of care is faulted for being rooted in class and culturally based
assumptions, supporting earlier contentions about the nature of schooling. Immigrant and
U.S.-born youth, however, both stated their preference for a second form of caring
steeped in “relations of reciprocity between teachers and students (Valenzuela, 1999;
cited in Antrop-Gonzalez & De Jesus, 2006:412). While aesthetically based caring tends
to set up barriers, authentic care appears to meet fundamental pre-condition for
promoting success through its focus on encouraging student participation in more
meaningful and personalized ways (Klem & Connel, 2004; cited in Antrop-Gonzalez &
De Jesus, 2006). Developing authentic forms of caring therefore not only promotes

41

academic rigor, but also draws youth into larger networks of pro-social adults through
more collaborative processes (Antrop-Gonzalez & De Jesus, 2006). A school enterprise
committed to building human social capital into the very process of schooling itself
reduces delinquency through the internalization of social norms and behaviours in a way
that allows students to feel more connected to their educational journey (Collier, 2006;
Enomito, 1997; Sergiovanni, 1994; Zyngier, 2003).
Organizational Models
Various writers (DeWit et al. 2002; Enomoto, 1997; Payne, Gottfredson &
Gottfredson, 2003; Sergiovanni, 1994) highlight the ‘school as community’ notion to
address the layering of factors from individual, family, school, and community that bear
on students’ success (Enomoto, 1997). Seriovanni (1994) distinguishes between two
operational modes of schools, arguing that ‘school as community’ differs from ‘school as
organization’ in the way that each concept anchors its work. ‘School as organization’
tends to focus on managerialism, placing attention on quality, productivity, efficiency,
and effectiveness. It works mainly from the standpoint of an economic model.
Sergiovanni (1994) states:
It is from economics, the parent of organizational theory,
that educational administration has borrowed its theories of
human nature and human motivation – theories built on the
simple premise that as human beings, we are motivated by
self-interest and thus seek to maximize our gains and cut
our losses (cited in Enomoto, 1997:215).
While the ‘school as organization’ analogy is said to occur in the midst of an
economically driven social environment where relationships are influenced by a
competitive logic based on rationality, status attainment, and achievement toward a

42

predefined set of goals (Enomoto, 1997), he argues not all people are motivated through
capitalist means.
‘School as community’, on the other hand, is founded on a less bureaucratic
framework where “collections of individuals…are bonded together by natural will
and…a set of shared ideas and ideals” (Enomoto, 1997: 218). This type of a social
environment contains a high sense of community, where “members know, care about,
and support one another, have common goals and sense of shared purpose, and to which
they actively contribute and feel personally committed” (Solomon, Battistich, Kim &
Watson, 1997, cited in Payne, Gottfredson & Gottfredson, 2003:236). This community
model relies on the development of social relationships based on shared kinship,
membership, locale, and principles that allow more students to gain access to the
opportunities made possible through supportive relationships with school staff (Collier,
2006). While organizationally run schools tend to impose their structure through
hierarchies, roles and role expectations, communally run schools tend to have more
inclusive cultures by engendering a shared set of values and goals (Collier, 2006;
Sergiovanni, 1994).
Though organizational and community models of education need not be a
mutually exclusive set of operational means, schools generally operate according to the
organizational model (Collier, 2006). By nature, organizational models establish climates
of risk; particularly for youth from positions of disadvantage. There is growing evidence
to suggest that as adults fail to model appropriate social interactions, teach an associated
set of social skills, and establish climates of care, antisocial behaviour among youth will
increase (Lewis, Sugai &Colvin, 1998). While most students can successfully navigate

43

the web of school dynamics (McCrystal, Percy & Higgins, 2007), students who do not
have the opportunity to participate in educational processes in ways that connect to their
realities are more likely to become excluded from opportunities afforded through the
educational system (DeWit, et al. 2002; Enomoto, 1997).
Zyngier (2003) emphasizes that school processes must acknowledge the presence
of inequity among their student bodies by going beyond programs and quick fix
solutions. Hirschi’s Social Control Theory maintains that schools can make a difference
in reducing delinquent behaviour by targeting youths’ need to belong. A positive
learning environment has already been cited as a key factor for developing healthy school
climate and fostering healthy belonging. Similar to home life, lack of structure and
consistency leads to a negative social learning environment, particularly when the social
climate of the classroom is poorly and inconsistently managed, and characterized by
conflict. Additional school factors that may create a climate of risk for detachment from
school differ by developmental level or grade. A Canadian study (Ma, 2003) involving
sixth and eighth grade students in New Brunswick schools found the single most
important school level factor for sixth graders was having clear expectations from
teachers. Academic press, understood as students’ belief about the importance of their
schoolwork, followed in order of importance. Grade eight students, however, ranked the
school’s disciplinary climate being of highest importance. Disciplinary climate referred
to perceptions about the fairness of rules, the enforcement of rules, students’ level of
participation in creating school rules, and staff responses to behaviour.
Viewing schools as a socializing structure for children and youth is not new. As
early as the 1960’s, Robins (1966) promoted the idea that schools should become

44

involved in raising children to fill in the gap from an apparent lack of parental discipline
in the home (cited in Gottfredson & Hirschi, 1990). Schools are sites where a major
portion of adolescent life occurs. As such, Gottfredson and Hirschi (1990) contend that
schools must have a net positive effect on youth through the cultivation of self control in
a way that leads to a crime free life. In their view, self control is predictable through
one’s commitment to mundane tasks at school, and expressed through activities, such as
completing homework, liking school, and doing well in school (Glueck & Glueck, 1950:
cited in Gottfredson & Hirschi, 1990). However, while having the opportunity to
facilitate success, schools may also impose strain on individuals depending on the extent
to which they are able to offer support and create positive places for learning. Schools
play an important role for addressing the underlying causes of substance use and abuse
among youth. Creating an environment that engenders attachment, induces commitment,
promotes involvement, and results in a shared set of beliefs, reduces the likelihood that
youth will develop behavioural trajectories involving delinquency and crime (Eith, 2004).
Longitudinal studies marking the direction of this relationship have been largely
absent (McLaughlin & Muncie, 2006). The problem with most research is in its’ use of
cross-sectional data that draws correlational results about substance use and attachment.
The extent to which substance use is influenced by students’ sense of attachment is not
yet identified in the literature, nor are we able to say which comes first – attachment or
substance abuse.

45

CHAPTER 4: METHODOLOGY
This section describes the theoretical orientation, method of data collection,
sample characteristics, and statistical procedure employed to test the research hypothesis.

Theoretical Orientation
Social Bonding Theory is particularly useful when seeking to discover those
aspects of society that inhibit deviant or non-conforming behaviour by increasing
compliance to social norms (McLaughlin & Muncie, 2006). Hirschi’s Social Bonding
Theory maintains that attachment to societal others, belief in the rules governing societal
order, commitment to conventional routes of achievement, and involvement in pro-social
activities are foundational to the development of pro-social behaviour (Hirschi, 2001;
cited in Hirschi, 1969).
Originally, Hirschi (1969) proposed that all social bonding inhibited deviance
(cited in Hawkins, Arthur & Catalano, 1995). Subsequent studies have outlined the
importance of reconsidering this idea based on a clearer articulation for the target of
attachment. Youth may develop social bonds to non-conforming individuals to influence
behaviour in undesirable ways. This was particularly critical when examining
relationships among peers (Hindelang, 1973 in Hoppe et al., 1998), and to parents
(Conger, 1976; in Hoppe et al. 1998). Attachment, therefore, is viewed as a positive
condition when the one to whom the individual is attached to is pro-social (Hawkins,
Arthur & Catalano, 1995). The same principle applies to institutions, such as schools. By

46

their nature, schools are considered to be conventional and therefore pro-social in nature
for the purposes of Hirschi’s notion of social bonding (Hawkins, Arthur & Catalano,
1995).
This study draws from previous work by utilizing two of Hirschi’s four social
bonding dimensions (attachment and involvement) by relating those elements to the
context of schools. The term school bonding was used to denote the presence of the two
elements: school attachment and school involvement. The first objective was to reexamine the relationship between school bonding and drug use. The second objective was
to examine the relationship between school bonding as a prior condition to drug use. The
two-part analysis was intended to not only assess the extent to which school bonding
items are able to predict current drug use, but also contribute to a new body of
information to assist in determining the extent to which school bonding predicts future
drug use. The findings are intended to guide the development of school policies for drug
use prevention programming and to facilitate positive youth development. Two research
questions guided the study:
1. Is students’ level of school bonding associated to current drug use?
2. Is students’ level of school bonding a preceding condition to drug use?

The Safe School Social Responsibility Survey for Secondary Students
The Safe School Social Responsibility Survey for Secondary Students (SSSRS) is
a self-report 285-item longitudinal survey for grade eight to twelve students (Appendix
A). This instrument measures student demographics, perceptions and experiences of
personal safety, substance use, school climate, self-concept, home life, social
responsibility, and community involvement. A privacy code allowed the researcher to
47

track participants from one year to the next while allowing students to remain
anonymous. The code was the product of five distinct questions that resulted in a unique
combination for each student comprised of alpha and numeric characters. The original
survey research project was completed in partnership with twelve B.C. school districts,
the BC Centre for Safe Schools and Communities, and the University of the Fraser
Valley. The purpose of the research was primarily intended to provide school districts
with comprehensive data for planning safe and socially responsible learning
environments.

Data Collection Process
The analysis for this study drew from the larger 2005/06 and 2006/07 SSSRS data
set of more than 30,000 grade eight to twelve students who participated in the survey
each year. From this larger sample, data from two medium-sized school districts formed
the starting point for the analysis, containing approximately 15,800 students. The two
districts, in particular, were chosen because they had participated in the survey research
for two consecutive years. Permission to use the data was granted by the faculty
researchers at the University of the Fraser Valley, as well as the research ethics board for
the University for secondary use of data.
The 2005/06 and 2006/07 administrations of the SSSRS were completed in the
spring of each year. Prior to survey date, a parent permission letter from the school
principal was sent to the homes of students using a passive consent process. Students
were informed ahead of time about the administration date at school through teacher
announcements, school websites, and newsletters. Parents, students, and school staff were
made aware of the survey’s intent, of its voluntary, confidential, and anonymous nature.
48

Classroom teachers received prior training by school administrators to standardize the
survey administration process.
Data entry occurred in the high security lab of the Centre for Criminal Justice
Research at the University of the Fraser Valley using a semi-automated process involving
the Scantron Cognition Enterprise system. Trained undergraduate university students
performed verification and correction of data.

Dependant Variable: Drug Use
The dependant variable of interest was drug use. The survey asks students to
provide information about the type and frequency of their drug use at school, and in the
community since September of the school year. Drug use referred to eight types of drugs:
marijuana, ecstasy, hallucinogens (LSD, acid), inhalants (glue, gas, aerosol), prescription
pills (not prescribed by a doctor), crystal meth, cocaine, and heroin (Section C, Questions
48 b-i). Alcohol use was excluded from the examination. The variable, drug use, was
operationalized by collapsing survey items to form a dichotomous variable indicating
whether (or not) a student had reported drug use from the eight types. Drug use became a
non-specific measure of drugs to test its relationship to school bonding. This allowed the
study to differentiate between students who either Used Drugs (indicated by answering
“Once or a few times”, “About once per month”, or “Every week or more”), or Did not
use Drugs (indicated by answering “Never” to any of the above listed drugs) since the
beginning of the school year.

49

Independent Variable: School Bonding
The independent variable used to examine the relationship to drug use drew from
two of the four social bonding elements included in Hirschi’s (1969) theory. The term
school bonding is used to denote the presence of attachment and involvement elements in
the context of schools. The following section outlines the rationale for operationalizing
school attachment and school involvement as the two major components of school
bonding in this study.

School Attachment
With respect to the school context, recent evidence has shown school attachment
to be the most robust indicator of the four social bonding elements (Eith, 2004). School
attachment involves a collection of singular relationships between the self and another, as
well as a relationship that students develop with ‘the school’ in a collective sense (Diaz,
2005; Hirschi, 1969, cited in Diaz, 2005; Maddox & Prinz, 2003). Various bodies of
research emphasize the multi-contextual and relational nature of school attachment that
guided the selection of three specific measures. The first measure of school bonding, “I
like school” (Survey Question 29) captured the broadest sense of school attachment. With
reference to its inverse relationship to delinquency, Hirschi stated “[t]he relation between
liking school and delinquency…is very strong” (1969:121). This general reference about
students’ relationship with school suggests that it goes beyond a collection of singular
relationships between the self and another, but one in which students view having with
‘the school’ in a collective sense (Eith, 2004; Maddox & Prinz, 2003). This also
acknowledged that students may have a certain fondness for the idea of school despite
their lack of enjoyment for it, or the specific school they attend (Diaz, 2005).
50

The next two measures reflected qualitative elements regarding students’
experiences found necessary to the development of attachment. The first , “Adults in my
school respect me” (Survey Question 35), captures positive interpersonal connections
(Anderman, 2003; Goodenow, 1992; Murray & Greenberg, 2000), making reference to
the importance of social support from school adults (Lewis, Sugai & Geoff, 1998;
Murray & Greenberg, 2000; Noddings, 1984; Scales, 2005). Goodenow specifically
refers to respect in his definition of school connectedness, stating it is “[t]he extent to
which [students] feel personally accepted, respected, included and supported by others in
the school social environment” (1992:3). This is echoed in Murray and Greenberg’s
(2000) work identifying that students who had poor relations with teachers, along with
weak bonds to school, scored lower on social adjustment in comparison to peers with
positive staff relations. Anderman (2003) maintained that as adults model respect to
students there is greater cohesion and sense of connectedness within the social
environment at school.
The third measure of school attachment, “My ideas and opinions are important to
at least one adult in my school” (Survey Question 23), allowed for the individual ego
level response through students’ feelings of being valued and acknowledged throughout
the day to day exchanges with school adults. This was supported by Hirschi’s (1969) and
Hunter and Danzker’s (2002; cited in Eith, 2004) articulation of attachment that refers to
a psychological mechanism linked to the conscience and superego. The item’s reference
to approval or validation by at least one adult at their current school captured students’
perceptions of their social position within a mainstream network. This was especially
important for at-risk students where school membership “…is not simply technical

51

enrolment, but students’ perceptions that others in the school, especially adults, are ‘for’
them, and that they count in the school” (Wehlage, 1989; cited in Goodenow, 1992:4).
Thus, school bonding through the application of attachment measures, refers to a
general and specific nature through which connections are developed and perceived
between the students, their teachers, and the institution. All three survey items were
measured on a 5-point Lickert scale that allowed for the following response categories:
“Strongly Disagree”, “Disagree”, “Undecided”, “Agree”, “Strongly Agree”.

School Involvement
School involvement is the most visible and tangible dimension of Hirschi’s four
social bonding elements (Eith, 2004). Within the school context, Hirschi (l969) provided
examples of involvement, such as regular school attendance, participating in a school
club or group, or watching a sports event (Siegel & McCormick, 1999). This is based on
the assumption that involvement protects youth from criminal behaviour, whereas lack of
involvement (or idleness) promotes it. Lack of involvement offers time and opportunity
for conceiving of, and getting into, trouble (Hirschi, 1969; Hunter & Dantzker, 2002;
cited in Eith, 2004), and for developing bonds with negative role models. School
involvement allows young people to be under the care and influence of pro-social adults
during developmental years (Eith, 2004). For the same reason that involvement in school
activities can function as a protective factor against the use of drugs, lack of involvement
presents a risk factor for engaging in drug use (Eaton, Brener & Kaan, 2008).
The first two measures of school involvement are specific behavioural items
representing the antithesis of regular school attendance. These include, “How often have
you skipped class?”(Question 21b) and “How often have you skipped all day? (Question
52

21c). Research has linked school absenteeism to negative social environmental factors in
school or home, as well as to poverty, and low academic motivation (Eaton, Brener &
Kann, 2008). School attachment and school involvement are interconnected ideas. If
students do not like school, they are more likely to have weak relations at school, and
therefore more prone to absenteeism because of a decreased desire to please adults
through behaviours that require school involvement (Hirschi, 1969; cited in Siegal &
McCormick, 1999). These two variables related to skipping behaviour reflected varying
degrees of absenteeism that were measured on a 5-point Lickert scale: “Never”; “At least
once this school year”; “About once per month”; “About once per week”; and, “More
than once per week”.
A third measure of school involvement, “How often have you participated in a
school group or club?” draws on voluntary behaviour. Capturing school participation
beyond a base level requirement of attending class suggested that students’ exposure to
delinquent opportunities or thoughts, and their proximity to positive adult influences
would decrease the chance of drug use involvement. This form of participation is also
cited in Hirschi’s example of behaviours constituting ‘legitimate’ activity (Hirschi, 1969),
and was captured in the survey on a five-point scale consisting of “Never”; “Once or
twice”; “Once a month”; “About once a week”; and, “More than once a week”.

Data Analysis
Assessing associations between two or more variables using bivariate analysis is a
basic yet useful method of data analysis (Sweet & Grace-Martin, 2003). To do so, the
Statistical Package for the Social Sciences (SPSS) has a basic statistical procedure for

53

exploring relationships among categorical variables through the Crosstab function
(Gravetter & Wallnau, 2002). Cross tabulations are a simple and effective way to
illustrate the joint distribution between two categorical variables at the nominal or ordinal
level of measurement by computing a grid that outlines all possible combinations of the
values for those variables (Frey, Botan & Kreps, 1999; Sweet & Grace-Martin, 2003).
The school bonding and drug use variables in this study were drawn from categorical
response sets using ordinal level increments. The drug use variable was re-coded into a
dichotomous variable to indicate whether (or not) a student had reported drug use from
among a variety of drug types listed in the survey. The Crosstab command was able to
provide information to indicate the strength and direction of the relationship between
school bonding variables and drug use.
In order to draw conclusions about the relationship and theoretical proposition
that school bonding was related to current, as well as the on-set of drug use, an estimate
of probability was used by conducting Chi Square tests of significance. Chi Square tests
explain the degree to which random chance is a likely explanation for a relationship
between variables (Sweet & Grace-Martin, 2003). They are also the appropriate test to
use in cases where numeric values are unavailable (Sweet & Grace-Martin, 2003), as was
the case for school bonding and drug use. Therefore, if each time the value of school
bonding showed a predictable change in the value of drug use, then the alternative
hypothesis proposing there would be a relationship between school bonding and drug use
should be real. The threshold used to indicate whether the relationship was statistically
significant, was established at the .05 level in keeping with statistical standards for the
social sciences (Sweet & Grace-Martin, 2003). In addition, when the relationship was at

54

or below the .10 level, conclusions about the observed relationship suggested that while
an actual relationship appeared to be the case, more research would be required before
drawing a more confident conclusion (Sweet & Grace-Martin, 2003). This process
allowed the study to make a determination about whether a significant enough difference
existed between students who used drugs and those who did not.
More specifically, the hypothesis that there would be a relationship between drug
use and social bonding to school was consistent with the literature. Therefore, to
determine whether school bonding was predictive of current and future drug use,
bivariate associations and Chi Square tests of significance were established for Phase 1
(Time 1 - 2005/06), as well as for Phase 2 (Time 1-2005/06 to Time 2-2006/07). For
Phase 2, the intent was to add a temporal component to the analysis by exerting
additional control over the drug use variable at Time 1 (2005/06). After first matching
student privacy codes for both years in order to track the on-set of drug use behaviour
from one year to the next, a sub-sample consisting of all students who reported no drug
use in Time 1 (2005/06) allowed the study to then conduct a re-examination of the
relationship between school bonding and drug use. This was used to determine whether
students who were less bonded to school in Time 1 (2005/06) were more likely to begin
using drugs in Time 2 (2007/08) over those who were more bonded to school. Social
bonding theory, and the construction of the school bonding variable led to the prediction
that an inverse relationship between drug use and social bonding to school would exist.
Therefore, the proposition that drug use at school, and in the community, both for the
current and following year, would be the same among students in relation to their level of
bonding to school was re-tested involving a temporal component.

55

Though bivariate analysis and significance testing forms part of a basic, yet
useful, process of determining the extent of the relationship between two variables,
allowing one to make accurate predictions about the other variable from a theoretical
basis (Gravetter & Wallnau, 2002), this method of prediction is not fool proof. This
statistical procedure allowed the study to answer a basic research question, while also
acknowledging the study to be exploratory in nature. Further study using greater
sophistication of analytic method would provide an even greater benefit.

Specifics about Sample Characteristics
The first phase of the data analysis (Time 1) consisted of examining a sample of
2,477 students from an annual joint district pool of approximately 15,800 students.
Matching all possible cases of students’ privacy code information found in Section A of
the SSSRS (Questions 1-5) across both years, resulted in a match rate of 15.6%. Match
failures occurred as a consequence of; 1) the voluntary nature of the privacy code
(students choosing not completing the first, second, or both years of privacy code
information); 2) students contributing privacy code information that was somehow
different from one year to the next, and; 3) the lack of available matches from grade
twelve participants in the 2005/06 year who will not have participated in the second year
of data collection due to completing school. Not surprisingly, the process of matching
student privacy codes reduced the sample substantially, and resulted in certain
characteristic differences between the unmatched group and matched group. Comparisons
of group differences are shown in Figures 1-4 for gender, grade, and ethnicity, and
educational aspiration.

56

Figure 1: Gender Comparison of Matched Sample to Unmatched Sample
50%

50%

57%

43%

Unmatched Sample
Matched Sample

Male

Female

Figure 1 indicates that gender was somewhat over-represented by females (57%)
over males (43%) in the matched sample in comparison to the even split for the
unmatched group.

Figure 2: Grade Comparison of Matched Sample to Unmatched Sample
25% 26%

26% 28%

25% 26%

24%
20%
Unmatched Sample
Matched Sample

Grade 8

Grade 9

Grade 10

Grade 11

]
Figure 2 shows similarity across grades eight to eleven. The greatest
inconsistency occurred among grade eleven students where the district percentage was
24% for the unmatched group, compared to 20% for the matched sample. As noted
above, grade twelve students from the 2005/06 year were excluded from this analysis.
Specifically, the matched sample consisted of 655 (or 26%) 8th graders, 693 (or 28%) 9th
graders, 643 (or 26%) 10th graders, and 486 (or 20%) 11th graders.

57

Figure 3: Grade Comparison of Matched Sample to Unmatched Sample

59%

63%
35%

Unmatched Sample

33%

Matched Sample
6%
Caucasian

Non Caucasian

4%

Aboriginal

By ethnicity, the greatest percentage difference existed in the 4% rise for
Caucasian participants among the matched sample. The matched sample consisted of
63% being of Caucasian descent, 4% Aboriginal, and 33% Non-Caucasian (excluding
Aboriginal). Group differences among the matched sample were also reflected in the Non
Caucasian (excluding Aboriginal) and Aboriginal categories, with a 2% drop in each.

Figure 4: Educational Aspiration Comparison between Matched Sample to Unmatched Sample
51%
35%
10%

41%
30%

Unmatched Sample
Matched Sample

14% 14%
5%

High School

Technical or
College

University
Degree

Post Graduate

Finally, after asking students, “What is the highest level of education that you
would like to complete?” (Question 17), figure 4 illustrates an elevated overall
commitment to higher education among the matched group. Comparing the matched
sample to the unmatched sample, 5% reported high school or less (vs. 10%); 30%
technical or college (vs. 35%); 14% university degree (vs 14%); and, 51% postgraduate

58

degree (vs. 41%). The most pronounced shift in educational differences lay at either end
of the aspiration spectrum, favouring higher aspirations among the matched sample.
Overall, when comparing the unmatched group (the general population of
students from the two school districts) to the matched group, there were slight differences
related to gender, grade, ethnicity, and educational aspirations. It is important to note that
the very nature of filling out student privacy codes may suggest that students in the
matched grouping are slightly more trusting of conventional institutions.
For the second phase of this analysis, a sub-sample of students who reported
“never” using drugs was drawn from the matched sample by further isolating all students
who reported “never” to using drugs in Time 1 (2005/06). The sample size was further
reduced by approximately 36% from the sample of 2,477 students, leaving 1578 cases to
make a determination about whether school bonding (involving school attachment, and
school involvement variables) predicted the initiation of drug use.

59

CHAPTER 5: KEY FINDINGS
This study examined the relationship between school bonding variables and drug
use. The association between variables was conducted in two phases. The results of the
first phase are highlighted in Tables 1 to 6 using bivariate analysis to assess the
association between school bonding (utilizing 3 school attachment, and 3 school
involvement variables) and drug use for the current year (Time 1 - 2005/06). Results for
the second phase of the analysis are highlighted in Tables 7-18, where bivariate analysis
was used to examine the association between school bonding (utilizing 3 school
attachment, and 3 school involvement variables) as a preceding condition (Time 1 –
2005/06) to drug use in Time 2 (2006/07).
For the most part, school bonding was inversely and significantly associated with
drug use for both the current and subsequent school year among high school youth. This
confirmed the predicted direction of the relationship for both objectives of the study
based on Hirschi’s Social Bonding Theory. The findings are presented in detail in the
following sections of this chapter.

60

Associations for School Bonding (Attachment Measures) and Drug Use
for Time 1 (2005/06 – current year)
School Attachment Dimensions
School Liking
Students were asked to respond to the statement, “I like school”. Table 1
demonstrates the inverse relationship between attachment and drug use behaviour.
Though moderately small, (r = -.197), participants showed significant differences in drug
use across all three measures of school attachment (p<.00l). This was most clearly
illustrated when comparing drug use between students who strongly disagreed, to
students who strongly agreed with the attachment statement, “I like school”. Among
those who strongly disagreed, 63% did not use drugs at school, compared to 38% who
did. In contrast, among those who strongly agreed, 92% did not use drugs at school,
compared to 8% who did. Students who had the lowest level of school liking were nearly
five times more likely to use drugs at school compared to those who had the highest level
of school liking.

Table 1: Differences between students who reported on whether or not they like school and

drug use at Time 1 (2005/06).
I like school

AT SCHOOL
Did Not
Used
Use Drugs
Drugs

IN THE COMMUNITY
Did Not
Used
Use Drugs
Drugs

Strongly Disagree
Disagree
Undecided
Agree
Strongly Agree

63%
73%
81%
86%
92%

48%
56%
62%
70%
81%

38%
27%
19%
14%
8%

Differences are significant at p<.001 level for both contexts.
* Questionnaire item #29 (see Appendix A) ** n=2477
61

52%
44%
39%
30%
19%

A similar pattern of association held true for drug use in the community (r =
-.189, p=.001). Among those who reported strong disagreement with the statement, “I
like school”, 48% did not use drugs in the community, while 52% did. In contrast, among
those who strongly agreed with the statement, 81% did not use drugs in the community,
while 19% did. Students who had the lowest level of liking school were nearly three
times more likely to use drugs in the community compared to those who had a high level
of school liking.
When comparing drug use across context (from school to community) by level of
school attachment, students were more likely to use drugs in the community than at
school. Among those who strongly disagreed with the statement, “I like school” drug use
was 38% at school compared to 52% in the community. By comparison, those who
strongly agreed with the statement, “I like school”, drug use was 8% at school compared
to 19% in the community.
Interestingly, and as might be expected, students who liked school and who also
reported using drugs, used them away from school in higher relative proportion than
students who did not like school. There was a greater rise in reported drug use from use at
school to use in the community as students’ liking for school increased; reporting a rise
of 138% across context among those who reported the highest level of liking, in
comparison to 37% among those who liked school the least.
Respect from Adults
A second measure of school attachment examined student’s perceptions of social
support from adults at school using responses to “Adults in my school respect me”.
Consistent with the predicted direction of this relationship, Table 2 illustrates the inverse

62

and relatively weak association (r = -.184, p=.001) between drug use at school and
feeling respected by adults at school. Among those who strongly disagreed with the
attachment statement, 53% did not use drugs at school, while 47% did. In contrast,
among those who strongly agreed with the statement, 89% did not use drugs, while 11%
did. Students who felt the least respected by adults at their school were four times more
likely to use drugs at school, compared to those who felt respected.

Table 2: Differences between students who reported on whether or not adults at their school

respected them and drug use at Time 1 (2005/06).
Adults in my school respect me

AT SCHOOL
Did Not
Used
Use Drugs
Drugs

IN THE COMMUNITY
Did Not
Used
Use Drugs
Drugs

Strongly Disagree
Disagree
Undecided
Agree
Strongly Agree

53%
68%
79%
86%
89%

37%
53%
61%
69%
79%

Differences are significant at p<.001 level.
* Questionnaire item #35 see Appendix A)

** n=2477

47%
32%
21%
14%
11%

63%
47%
39%
31%
21%

The same inverse pattern of drug use in relation to level of attachment held true
for drug use in the community (r = -.178, p=.001). Among those who reported strong
disagreement with the statement, “Adults in my school respect me”, 37% did not use
drugs, while 63% did. Among those who reported strong agreement, 79% did not use
drugs, while 21% did. Students who felt the least respected by adults at their school were
three times more likely to use drugs in the community, compared to those who felt a high
level of respect.

63

Once again, when comparing drug use across context (from school to community)
there was a greater rise in drug use, from use at school, to use in the community. Those
who felt the most respected by adults showed a relative rise of 91% across context, in
comparison to 34% from among students who felt the least amount of respect.
Ideas & Opinions
The third measure of school attachment for the Time 1 (2005/06) analysis
examined students responses to the statement, “My ideas and opinions are important to at
least one adult in my school”. Again, consistent with prediction, it resulted in the weakest
inverse relationship of the three attachment measures (r=-.069, p=<.001) for the current
year. Table 3 illustrates that among those who strongly disagreed with the attachment
statement, 65% did not use drugs at school, while 47% did. In contrast, among those who
strongly agreed with the statement, 82% did not use drugs, while 18% did. Students who
strongly disagreed with the statement that their ideas and opinions were important to at
least one adult at their school were more than twice as likely to use drugs at school,
compared to those who felt strongly that their ideas and opinions mattered.

64

Table 3: Attachment: Differences between students who reported on whether or not they felt

their ideas and opinions were important to at least one adult in their school and
drug use at Time 1 (2005/06).
My ideas and opinions are important
to at least one adult in my school.

AT SCHOOL
Did Not
Used
Use Drugs
Drugs

IN THE COMMUNITY
Did Not
Used
Use Drugs
Drugs

Strongly Disagree
Disagree
Undecided
Agree
Strongly Agree

65%
71%
83%
82%
84%

50%
52%
67%
65%
71%

35%
29%
17%
18%
16%

51%
48%
33%
36%
29%

Differences for school and community are significant at p<.001 level.
* Questionnaire item #23 (see Appendix A) ** n=2477

A similar pattern held true for drug use in the community (r = -.079, p=.001).
Among those who reported strong disagreement with the statement, “My ideas and
opinions are important to at least one adult in my school”, 50% did not use drugs in the
community, while 51% did. In contrast, among those who strongly agreed with the
statement, 71% did not use drugs in the community, while 29% did. Students who
strongly disagreed that their ideas and opinions were important to at least one adult at
their school were almost two times more likely to report drug use in the community over
those who strongly agreed that their ideas and opinions mattered.
Finally, when comparing drug use across context (from school to community)
with respect to students’ feeling valued for their ideas and opinions, there was a greater
relative rise in drug use, from use at school, to use in the community, among students
who felt valued for their ideas and opinions from school adults. This was illustrated in a
reported rise of 81% across context, in comparison to only 37% from among students

65

who did not feel their ideas and opinions were of high importance to adults at their
school.
This school attachment section for Time 1 highlights three general patterns of
association for school attachment and drug use in Year 1 (2005/06). First, students with
lower levels of attachment to school, measured by students’ liking for school, feeling
their ideas and opinions were important to at least one adult at school, and perceptions
that adults at school respected them, were more likely to use drugs at school.
Accordingly, among those who used drugs, the greatest difference in reported drug use
occurred between those who were highly attached to school (indicated by ‘strongly
agree’) and those who were very unattached (indicated by ‘strongly disagree’). Second,
the same general pattern of association held true for drug use in the community though
the context of the school appeared to moderate drug use. Across all levels of school
attachment, drug use was more likely to occur in the community. Third, students who
were more socially bonded to school appeared to be more discerning in their decisionmaking to use drugs away from school. This last finding, in particular, highlights the
need for further research to gain a better understanding for the extent to which trends
about frequency of use and type of use differ between mainstream youth and those
showing higher levels of disconnectedness from school. Overall, the school attachment
dimension of school bonding for Time 1 (2005/06) indicated that all three measures
ranged from being weak to moderate in strength, though highly significant in their
association to drug use at school, and slightly less so for drug use in the community.
More specifically, school liking, closely followed by respect from adults, were the most
robust predictors of current drug use at school among high school students.

66

School Involvement Dimensions
The following section highlights the results for school involvement in relation to
drug use for the current year (2005/06). These are featured in Tables 4 to 6.
Skipped Class
Students were asked to indicate how often they skipped class. Table 4 illustrates a
moderately strong and positive relationship (r = .373, p=.001) between skipping class and
drug use. Among those who skipped class “More than once per week”, 33% did not use
drugs at school, while 68% did. In contrast, among those who “Never” skipped a class
94% did not use drugs at school, while 6% did. Students who skipped class “more than
once per week, were at least eleven times more likely to use drugs at school compared to
those who never skipped class.

Table 4: Differences between students who reported on whether or not they skipped a class

and drug use at Time 1 (2005/06).
Skipped a class

AT SCHOOL
Did Not
Used
Use Drugs
Drugs

IN THE COMMUNITY
Did Not
Used
Use Drugs
Drugs

Never
At least once this school year
About once per month
About once per week
More than once per week

94%
82%
64%
61%
33%

85%
61%
39%
31%
16%

6%
18%
36%
39%
68%

15%
39%
61%
69%
84%

Differences for school and community are significant at p<.001 level.
* Questionnaire item #21b (see Appendix A) ** n=2477

A similar and even stronger relationship held true for skipping class and drug use
behaviour in the community (r = .449, p=.001). Among those who skipped class “More
than once per week”, 16% did not use drugs in the community, while 84% did. In
contrast, among those who “Never” skipped class, 85% did not use drugs in the
67

community, while 15% did. Students who skipped class more than once per week were
nearly six times more likely to use drugs in the community, compared to those who never
skipped class.
When comparing drug use across context (from school to community) with
respect to students skipping class behaviours, there was a greater relative rise in drug use,
from use at school, to use in the community, among students who more involved in
school. A reported rise of 150% across context occurred among those most involved, in
comparison to a more modest 23% rise in reported use among those least involved.
Skipped all Day
Table 5 similarly outlines a moderately strong and positive relationship between
the second negative indicator of school involvement, “skipping all day” and drug use at
school (r = .331, p=.001). Among those who skipped all day “More than once per week”,
16% did not use drugs at school, while 84% did. In comparison, among those who
“never” skipped all day, 89% did not use drugs at school, while 11% did. Students who
skipped a day more than once per week were almost eight times more likely to use drugs
at school compared to those who never skipped a day.

68

Table 5: Differences between students who reported on whether or not they skipped all day

and drug use at Time 1 (2005/06).
Skipped all day

AT SCHOOL
Did Not
Used
Use Drugs
Drugs

IN THE COMMUNITY
Did Not
Used
Use Drugs
Drugs

Never
At least once this school year
About once per month
About once per week
More than once per week

89%
70%
53%
44%
16%

75%
47%
31%
19%
12%

11%
30%
47%
56%
84%

25%
53%
69%
82%
88%

Differences are significant at p<.001 level.
* Questionnaire item #21c. (Appendix A) ** n=2477

A similar positive relationship held true for ‘skipping all day’ behaviour and drug
use in the community (r = .341, p=.001). Among those who skipped all day at the rate of
“More than once a week”, 12% did not use drugs in the community, while 88% did. In
contrast, among those who “Never” skipped all day, 75% did not use drugs in the
community, while 25% did. Students who skipped all day were nearly four times more
likely to use drugs at school compared to those who never skipped a day.
Once again, when comparing drug use across context with respect to students’
skipping all day behaviours, there was a greater relative rise toward drug use in the
community among students who had the highest level of involvement (175%) in
comparison to those least involved (5%).
School Club or Group Participation
The third measure of involvement examined students’ responses to the statement,
“How often have you participated in a school club or group”? Compared to skipping
behaviours, Table 6 demonstrates a weaker, though still a significant inverse relationship

69

(r = -.135, p=.001) between school club or group participation and drug use at school.
Among those who said they ‘never’ participated, 74% did not use drugs at school, while
26% did. In contrast, among those who said they participated “More than once per week,
89% did not use drugs at school, while 12% did. Students who were involved in school
clubs or groups more than once per week were two times more likely to use drugs at
school compared to those who were not involved in this way.

Table 6: Differences between students who reported on whether or not they participated in a

school club or group and drug use at Time 1 (2005/06).
Participated in a school club or group

AT SCHOOL
Did Not
Used
Use Drugs
Drugs

IN THE COMMUNITY
Did Not
Used
Use Drugs
Drugs

Never
Once or twice
Once a month
About once a week
More than once per week

74%
84%
84%
85%
89%

57%
66%
67%
69%
77%

26%
17%
16%
15%
12%

43%
34%
33%
31%
24%

Differences are significant at p<.001 level.
* Questionnaire item #71a. (Appendix A)
** n=2477

A similar pattern held true for drug use in the community (r = -.132, p=.001).
Among those who ‘never’ participated in a school club or group, 57% did not use drugs
in the community, compared to 43% who did. In contrast, among those who participated
“More than once per week”, 77% did not use drugs in the community, while 24% did.
Students who were involved in school clubs or groups more than once per week were less
than half as likely to use drugs in the community compared to those who were never
involved in school clubs or groups.

70

When comparing drug use across context (from school to community) with
respect to students participation in school clubs or groups, the linear pattern of the
relationship does not follow the evenly stepped pattern of increase and decline that was
observed in the previous two indicators of school involvement. Though generally similar
in its direction, the greatest increase in drug use across context by level of involvement
was between those who never participated at all (showing a rise of 65%), and those who
participated minimally at the rate of about once per year (showing a rise of 100%).
This school involvement section for Time 1 highlights four patterns of association
for school involvement and drug use for the current year (Time 1-2005/06). First, school
involvement was inversely associated to drug use. Students who reported higher levels of
school involvement measured by participating in a school club or group, and through
non-participating forms of behaviour involving skipping class, and skipping all day were
less likely to use drugs at school. Recalling that two of the three indicators of school
involvement dimensions are negative measures of involvement, the tables correctly
indicate positive relationships for non-participating behaviours due to their reverse-coded
nature. Second, the same general pattern of association holds true for drug use in the
community. Third, across all reported levels of school involvement, drug use was more
likely to occur in the community. Fourth, students who were more involved at school
were also more likely to report higher increases in drug use behaviour by context when
comparing their drug use behaviour at school to drug use behaviour in the community.
While general trends for school involvement were similar to those for school attachment,
the school involvement dimension for Time 1 (2005/06) indicates that non-participating
behaviours (class skipping, skipping all day) were even more powerful indicators of drug

71

use than were indicators of attachment. Participating in clubs or groups was weaker in
comparison to skipping behaviours that were moderate to strong. For both dimensions of
school bonding (school attachment, and school involvement), the predicted direction of
the research hypothesis was confirmed, supporting the notion that school bonding is
associated to current drug use.

Associations for School Bonding in Time 1 (2005/06) and Subsequent
Drug Use in Time 2 (2006/07)

The previous section has highlighted measures of association for Time 1
(2005/06) between drug use and school bonding, using school attachment and school
involvement variables. The next section will highlight the results of school bonding
variables to determine whether school bonding is associated to drug use in the following
year (Time 2 – 2006/07). The results of this one-year tracking process are presented in
Tables 7-18.

School Attachment Dimensions
School Liking
The first two tables of this section outline the inverse relationship for the first
indicator of school attachment, “I like school” for Time 1 (2005/06) and subsequent drug
use at school (r = -.092, p=.01), and in the community (r = -.075, p=.05) for the following
school year (Time 2 – 2006/07).

72

Table 7: Differences between students who reported on whether or not they liked school at

Time (2005/06) and drug use at school at Time 2 (2006/07).
I like school

Did Not
Use drugs

Used
Drugs

Strongly Disagree
Disagree
Undecided
Agree
Strongly Agree

84%
89%
88%
92%
95%

16%
11%
12%
9%
5%

*Difference is significant at p<.01 level.
** Questionnaire item #29 (see Appendix A)
*** n=1578

For drug use at school, Table 7 shows that among those who strongly disagreed
with the statement, “I like school”, 84% did not use drugs in the following school year,
while 16% did. In contrast, among those who strongly agreed with the statement, 95%
did not use drugs in the following school year, while only 5% did. Students who had the
lowest level of liking school were three times more likely to use drugs at school in the
following year than those who had a high level of school liking.
Similarly, for drug use in the community, Table 8 shows the inverse relationship
between the statement, “I like school” and drug use in the following year. However, in
this context, students who had the lowest level of school liking were almost twice as
likely to use drugs in the community one year later compared to students who had the
highest level of school liking.

73

Table 8: Differences between students who reported on whether or not they liked school at

Time (2005/06) and drug use in the community at Time 2 (2006/07).
I like school

Did Not
Use Drugs

Used
Drugs

Strongly Disagree
Disagree
Undecided
Agree
Strongly Agree

74%
76%
76%
79%
87%

26%
24%
24%
22%
14%

Difference is significant at p<.05 level.
* Questionnaire item #29 (see Appendix A) ** n=1578

Cross-context use of drugs, from school to community showed that students were
more likely to use drugs in the community. Among those who strongly disagreed with the
statement, “I like school” drug use was 16% at school compared to 26% in the
community. By comparison, those who strongly agreed with the statement, drug use was
5% at school compared to 14% in the community.
Respect from Adults
The following two tables outline the inverse relationship between the second
indicator of school attachment, “Adults in my school respect me” for Time 1 (2005/06)
and subsequent drug use at school (r = -.076, p=.001), and in the community (r = -.077,
p=.05) for Time 2 (2006/07).

74

Table 9: Differences between students who reported on whether or not they felt adults in

their school respected them at Time 1 (2005/06) and drug use at school at Time
2 (2006/07).
Adults in my school respect me

Did Not
Use Drugs

Used
Drugs

Strongly Disagree
Disagree
Undecided
Agree
Strongly Agree

73%
87%
89%
91%
93%

27%
13%
11%
9%
7%

Difference is significant at p<.05 level.
* Questionnaire item #35 (see Appendix A) ** n=1578

For drug use at school, Table 9 shows that among those who strongly disagreed
with the statement, “Adults in my school respect me”, 73% did not use drugs in the
following school year compared to 27% who did. In contrast, among those who strongly
agreed with the statement, 93% did not use drugs in the following school year compared
to 7% who did. Students who felt the least respected by adults at their school were nearly
four times more likely to use drugs at school in the following school year than those who
felt the most respected.
Similarly, for drug use in the community, Table 10 also shows the inverse
relationship between the statement, “Adults at my school respect me” and subsequent
drug use. However, in this context, students who felt the least respected by adults at their
school were more than twice as likely to use drugs in the community in the following
year, compared to students who felt the most respected.

75

Table 10: Differences between students who reported on whether or not they felt adults in

their school respected them at Time 1 (2005/06) and drug use in the community
at Time 2 (2006/07).
Adults in my school respect me

Did Not
Use Drugs

Used
Drugs

Strongly Disagree
Disagree
Undecided
Agree
Strongly Agree

66%
75%
76%
79%
84%

34%
26%
24%
21%
16%

Difference is significant at p<.05 level.
* Questionnaire item #35 (see Appendix A) **n=1578

Cross-context use of drugs, from use at school, to use in the community, showed
that students were more likely to use drugs in the community. Among those who strongly
disagreed with the statement, “Adults in my school respect me” in Time 1 (2005/06),
drug use was 27% at school compared to 34% in the community. By comparison, those
who strongly agreed with the statement, drug use was 7% at school compared to 16% in
the community.
Ideas and Opinions
The following two tables outline the inverse relationship between the third
indicator of school attachment, “My ideas and opinions are important to at least one adult
in my school” for Time 1 (2005/06) and subsequent drug use at school (r = -.076, p=.01),
and in the community (r = -.063, p=.01) for Time 2 (2006/07).

76

Table 11: Differences between students who reported on whether or not they felt their ideas

and opinions were important to at least one adult in their school at Time 1
(2005/06) and drug use at school at Time 2 (2006/07).
My ideas and opinions are important
to at least one adult in my school.

Did Not
Use drugs

Used
Drugs

Strongly Disagree
Disagree
Undecided
Agree
Strongly Agree

82%
80%
90%
91%
93%

18%
20%
10%
9%
8%

Difference is significant at p<.01 level.
*Questionnaire item #23 (see Appendix A) **n=1578

For drug use at school, Table 11 shows that among those who strongly disagreed
with the statement, “My ideas and opinions are important to at least one adult in my
school”, 82% did not use drugs the following school year compared to 18% who did. In
contrast, among those who strongly agreed with the statement, 93% did not use drugs in
the following school year compared to 8% who did. Students who strongly felt their ideas
and opinions were not important to at least one adult at school were more than twice as
likely to use drugs in the following school year, compared to those who strongly agreed
that their views mattered.
Similarly for drug use in the community, Table 12 shows the inverse relationship
between students reporting on “My ideas and opinions are important to at least one adult
in my school” in Time 1 (2005/06) and subsequent drug use in the community. However,
in this context, students who strongly felt their ideas and opinions were not important to
at least one adult at school were more than twice as likely to use drugs in the community

77

in the following school year, compared to students who strongly felt their ideas and
opinions mattered.

Table 12: Differences between students who reported on whether or not they felt their ideas

and opinions were important to at least one adult in their school at Time 1
(2005/06) and drug use in the community at Time 2 (2006/07).
My ideas and opinions are important
to at least one adult in my school.

Did Not
Use drugs

Used
Drugs

Strongly Disagree
Disagree
Undecided
Agree
Strongly Agree

61%
72%
79%
78%
83%

39%
28%
21%
22%
17%

Difference is significant at p<.01 level.
*Questionnaire item #23 (see Appendix A) **n=1578

Cross-context drug use from school to community showed that students are more
likely to use drugs in the community. Among those who strongly disagreed with the
statement, “My ideas and opinions are important to at least one adult in my school”, drug
use was 18% at school compared to 39% in the community. By comparison, those who
strongly agreed with the statement, drug use was 8% at school compared to 17% in the
community.
This school attachment section for Time 1 to Time 2 highlights two general
patterns in relation to drug use among students in the following year. First, all three
measures of school attachment were weak, though significant, in their inverse
relationship to subsequent drug use behaviour. Decreased levels of attachment to school,
indicated by individuals’ liking for school, respect from adults at school, and perceptions
that their ideas and opinions were important to at least one school adult, were linked to a
rise in reported drug use for the following school year. Second, as was true for the Time 1
78

(2005/06) analysis, drug use was consistently higher in the community than at school
across all levels of school attachment.

School Involvement Dimensions
The final six tables outline the relationship between the three behavioural
variables representing school involvement for Time 1 (2005/06) and drug use for the
following school year (Time 2 – 2006/07).
The following two tables outline the positive relationship between the first
negative measure of school involvement, “Skipped a class” for 2006 and subsequent drug
use at school (r = .099, p= .001), and in the community (r = .187, p=.001). Again, while
positive in their association to the reverse-coded variable, they are consistent with the
predicted inverse direction of the relationship between school bonding and drug use.

Table 13: Differences between students who reported on whether or not they skipped a class

at Time 1 (2005/06) and drug use at school at Time 2 (2006/07).

Skipped a class

Did Not
Use Drugs

Used
Drugs

Never
Once this school year
Once per month
Once per week
More than once per week

92%
90%
83%
77%
80%

8%
10%
17%
23%
20%

Difference is significant at p<.001 level.
* Questionnaire item #21b (see Appendix A) * n=1578

For drug use at school, Table 13 shows that among those who skipped a class
“More than once per week”, 80% did not use drugs in the following school year
compared to 20% who did. In contrast, among those who “Never” skipped a class, 92%
79

did not use drugs in the following year compared to 8% who did. Students who skipped
class more than once per week were two and a half times more likely to use drugs at
school in the following year, compared to students who never skipped a class.
Similarly, for drug use in the community, Table 14 also shows the positive
relationship between “skipping a class” in Time 1 (2005/06) and subsequent drug use in
the community in Time 2 (2006/07). However, in this context, students who skipped class
more than once per week were nearly four times more likely to use drugs in the
community in the following year, compared to students who never skipped a class.

Table 14: Differences between students who reported on whether or not they skipped a class

at Time 1 (2005/06) and drug use in the community at Time 2 (2006/07).
Skipped a class

Did Not
Use Drugs

Used
Drugs

Never
Once this school year
Once per month
Once per week
More than once per week

84%
72%
63%
70%
40%

16%
28%
37%
30%
60%

Difference is significant at p<.001 level.
Questionnaire item #21b (see Appendix A) n=1578

Cross-context use of drugs, from school to community shows that students were
more likely to use drugs in the community. Among those who skipped class more than
once per week, drug use was 20% at school compared to 60% in the community. By
comparison, those who never skipped class, drug use was 8% compared to 16% in the
community.

80

Skipped all Day
The following two tables outline the positive relationship between the second
negative measure of school involvement, “Skipped all day” for Time 1 (2005/06) and
subsequent drug use at school (r = .072, p= .05), and in the community (r = .120, p= .001)
for Time 2 (2006/07).

Table 15: Differences between students who reported on whether or not they skipped all day

at Time 1 (2005/06) and drug use at school at Time 2 (2006/07).

Skipped all day

Did Not
Use Drugs

Used
Drugs

Never
Once this school year
Once per month
Once per week
More than once per week

91%
88%
83%
80%
50%

9%
12%
18%
20%
50%

Difference is significant at p<.05 level.
* Questionnaire item #21c (see Appendix A) * n=1578

For drug use at school, Table 15 shows that among those who skipped all day
“More than once per week”, similar numbers were reported between those who used
drugs and those who did not. However, among those who “Never” skipped all day, 91%
did not use drugs one year later compared to 9% who did. Students who skipped all day
more than once per week were almost six times more likely to use drugs at school one
year later than those who skipped all day more than once per week.
Similarly, for drug use in the community, Table 16 shows the positive relationship
between those who “Skipped all day” in Time 1 (2005/06) and subsequent drug use in the
community for the following school year. However, in this context, students who skipped

81

a day the most were five times more likely to use drugs in the community in the
following year, compared to students who never skipped a day.

Table 16: Differences between students who reported on whether or not they skipped all day

at Time 1 (2005/06) and drug use in the community at Time 2 (2006/07).

Skipped all day

Did Not
Use Drugs

Used
Drugs

Never
Once this school year
Once per month
Once per week
More than once per week

80%
70%
64%
80%
-

20%
30%
37%
20%
100%

Difference is significant at p<.001 level.
Questionnaire item #21c (see Appendix A)

n=1578

Cross-context use of drugs, from use at school to use in the community, showed
students being more likely to use drugs in the community. Among those who skipped a
day more than once per week, drug use was 50% at school and 100% in the community.
By comparison, among those who never skipped a day, drug use was 9%, compared to
20% in the community.
School Club or Group
The final two tables outline the inverse relationship between the third indicator of
school involvement, “Participated in a school club or group” for Time 1 (2005/06) and
subsequent drug use at school (r = -.094, p < .001), and in the community (r = .060, p<
.067 for Time 2 (2006/07).

82

Table 17: Differences between students who reported on whether or not they participated in

a school club or group at Time 1 (2005/06) and drug use in school at Time 2
(2006/07).
Participated in a school club or group

Did Not
Use Drugs

Used
Drugs

Never
Once this school year
Once per month
Once per week
More than once per week

87%
90%
87%
93%
96%

13%
10%
13%
7%
4%

Difference is significant at p<.001 level.
Questionnaire item #71a (see Appendix A)

n=1578

For drug use at school, Table 17 shows that among those who “Never”
participated in a school club or group” 87% did not use drugs in the following school
year while 13% did. In contrast, among those who participated “More than once per
week”, 96% did not use drugs one year later compared to 4% who did. Students who
never participated in school clubs or groups were at least three times more likely to use
drugs at school the following year than those who participated “More than once per
week”. While the inverse pattern was consistent with general findings for this study at
either end of the school involvement scale, mid level associations did not follow the
regular stepped pattern of increase and decline relative to drug use.
For drug use in the community, Table 18 illustrates only a marginally related association
between participation in a school club or group and subsequent drug use in the community
(r = -.097, p <.067). Though the Chi Square tests of significance did not meet the .05 level of
stringency to determine whether the relationship was due to random chance, the same general
pattern of direction is there, as in the case for the skipping class, and for skipping all day
behaviours.

83

Table 18: Differences between non-drug using students who report on whether or not they

participated in a school club or group at Time 1 (2005/06) and drug use in the
community at Time 2 (2006/07).
Participated in a school club or group

Did Not
Use Drugs

Used
Drugs

Never
Once this school year
Once per month
Once per week
More than once per week

75%
79%
74%
82%
83%

25%
21%
26%
18%
17%

Difference not significant at .05
Questionnaire item #23 (see Appendix A)

n=1578

In this final item, cross-context drug use from school to community also showed that
students were more likely to use drugs in the community over school. Among those who never
participated in school clubs or groups, drug use was 13% at school compared to 25% in the
community. By comparison, those who participated more than once per week, reported drug use
at 4% at school, compared to 17% in the community.
This school involvement section for the time series examination indicated that
school involvement was inversely and significantly related to subsequent drug use. When
comparing the strength of their relationship to the first phase of the analysis for Time 1
(2005/06), the magnitude of these relationships was lower as a whole. This drop in
magnitude may indicate the absence of drug use effects on students’ school bond by
having controlled for drug use in year 1 to explore the occurrence of a directional element
for the relationship. Though covariability between school bonding and drug use was
weak in this examination, the predicted direction of the hypothesis, stating that there
would be a difference in the distribution of drug use among categories of students by
level of social bonding to school, was confirmed.

84

CHAPTER 6: DISCUSSION AND CONCLUSION
This study adopted a school bonding perspective based on Hirschi’s Social
Bonding Theory utilizing attachment and involvement dimensions for its application to
the school context. The intent was to better understand the potential role of schools in
promoting positive youth development by examining the extent to which specific school
factors were associated to drug use. As a derivative of Social Bonding Theory, school
bonding was applied at the nexus of the individual and the school, acknowledging
schools as a primary socializing institution for youth. The study had two primary
objectives: to confirm the relationship between school bonding and current drug use,
using cross-sectional data for Time 1 (2005/06); and, then to elaborate on this
relationship with a view to examine school bonding as a preceding condition to drug use
using longitudinal data from the Safe School Social Responsibility Survey for Secondary
Students. The theoretical and practical implications for school bonding and its association
to drug use are summarized below.

Objective 1
The findings confirmed an inverse relationship between current drug use and
school bonding. These findings support Hirschi’s Social Bonding Theory that argues
delinquency is more likely to occur when the individual’s bonds to society are weak or
broken (Hirschi, 1969; cited in Moyer, 2001). Contrary to previous research (Eith, 2004)
identifying school attachment as the most robust element of the four social bonding

85

elements , this study found school involvement to be the more powerful predictor of the
two elements under study.

School Attachment
School attachment showed a predictable increase in drug use as students’ level of
school attachment decreased. More specifically, lower levels of school attachment, using
three items that included school liking, feeling respect from adults at school, and feeling
as though one’s ideas and opinions were important, resulted in more students reporting
current drug use at school and in the community. Of the three attachment measures, both
a general liking for school and students feelings of respect were the stronger of the three
measures, and similar in their weak to moderate inverse association found for drug use at
school (r = -.197 and r=.184 respectively), and for drug use in the community (r = -.189
and r=.178 respectively). Students who liked school the least were three times more
likely to use drugs at school, compared to those who liked it the most. Students who felt
least respected by adults in their school were nearly four times more likely to use drugs at
school, compared to those who felt a high degree of respect.
These findings re-affirm the constructive and protective role of schools, including
the need for caring interpersonal relationships. Past research has already indicated that
higher rates of delinquency occur among students who have less positive psychological
orientations to the school (Eith, 2004). These findings also substantiate Hirschi who
stated that students who were not attached to school would have a general dis-liking for it
(Diaz, 2005), and be more likely to commit acts of delinquency (including drug use), if
connections to adult support systems were weak or broken (Hirschi, 1969).

86

Slightly more than half of the grade eight to eleven students in the sample were
attached to school through positive or pro-social connections with the adults, and
approximately two-thirds of students said they liked school. These findings suggest that
many students are not attached through positive or pro-social relations at school, and
therefore more vulnerable to delinquency. Among the 23% of students who did not like
school, about half also did not have positive or pro-social connections with adults at their
school. Further analysis would be required to understand how this may be related to
grade, gender, or other demographic variables of interest.

Involvement
While school attachment provided a psychological context for understanding drug
use, school involvement linked school bonding directly to behavioural measures. First,
the majority of students (72%) never skipped school for an entire day. However, this was
not the case for skipping a class where slightly more than half of the students reported
skipping class with some regularity (once per month or more). Of those, 11% skipped
class once per week or more. In its association to drug use, there were stepped linear
increases in drug use as school bonding decreased. This appears to be important when
considering the moderate to strong association between school absenteeism and drug use.
While all three measures of school involvement (skipping all day, skipping a class,
participating in a school club or group) were significant, skipping class was the most
powerful predictor of drug use at school (r=.373, p<.001), as well as for drug use in the
community (r=.449, p<.001). This was followed by ‘skipping all day’ where the
association was more moderate for drug use at school (r=.331, p<.001), and for drug use
in the community (r=.341, p<.001). Students’ lack of involvement in pro-social school
87

activities predicted drug use in the following school year even more powerfully than
school attachment indicators.

Objective 2
While this study was unable to determine whether low school bonding caused
drug use, or whether drug use contributed to low school bonding, there is reason to
believe in their influencing effect. This portion of the analysis provides initial evidence
that school bonding is a preceding condition to drug use for both school attachment and
school involvement measures. This occurred by examining among a sub population of
students who identified that they had not used drugs in Time 1 (2005/06), and then reexamining the strength and direction of the association between school bonding for Time
1 (2005/06) and any subsequent drug use behaviour (in Time 2 - 2006/07).

School Attachment
Concerning school attachment as a preceding condition to drug use, all three
measures of school attachment (school liking, respect from adults, and student
perceptions that their opinions and ideas were important) showed a relationship to
subsequent drug use. While weak in magnitude, school liking was the most robust
predictor of drug use at school (r = -.092, p=<.01), and drug use in the community (r = .075, p=<.05) of the three. In addition, cell counts for the latter two measures limited the
study from drawing conclusions that were more confident. To do so, further research
using larger sample sizes would be required.

88

Involvement
For school involvement measures, early indications showed that all three
measures of school involvement (skipping class, skipping all day, participating in school
club or group) predict drug use in the following year. However, the only measure that had
a sufficiently large cell count for confirming a chi square level of significance was for the
variable ‘participating in a club or group’. In this instance, there was a weak, though
significant relationship for drug use at school (r = -.097, p = <.001). For drug use in the
community the level of significance fell outside the .05 level required for the social
sciences (r = -.062, p = <.067) (Gravetter & Wallnau, 2002). Concerning skipping
behaviours across time, given that the relationship fell in the same expected direction for
drug use in either context (school or community), and that the strength of the direction for
the Time 1 analysis was powerful and significant, it is reasonable to expect that if a larger
sample size had been available, a more confident conclusion concerning this relationship
might be made.
Concerning the second objective of the study, it was interesting to note that the
associations between school bonding, both for school attachment and school involvement
measures, were weaker overall in their relationship to subsequent drug use. This suggests
the presence of a possible effect from having controlled for the reciprocal and reinforcing
negative effect of drug use on school bonding by having isolated the sub-sample of
students who had not used drugs in Time 1 (2005/06) for the analysis. This was noted
when comparing the attachment measure of school liking between Time 1 (2005/06) and
Time 2 (r=-.197 vs. r=-.092 respectively). This was also the case for school involvement.
For example, skipping class behaviour for Time 1 showed a similar trend between the
Time 1 to Time 2 analysis (r =.373 vs. r = .115 respectively). Taken together, these
89

patterns reinforce the idea that early protection to avoid and interrupt the compounding
effect of risk from drug use is important. When students are not embedded in cohesive
and supportive pro-social networks, the magnification of risk from potential negative
peer, family, school and community factors may allow problems to gain momentum in
students’ lives (Health Canada, 2001).
The results for both objectives also confirmed the existence of a contextual effect
related to school bonding and drug use. First, drug use was higher in the community
overall. Second, students who were less bonded to school appeared also to be less
discerning about the setting in which they used drugs, whether drug use occurred at
school or in the community. Social Bonding Theory was tested as an independent
variable linked to drug use, based on the assumption that schools are conventional sites
where youth are under the care and supervision of adults, and presupposed that students’
time was predominantly occupied by involvement in learning tasks (Hawkins & Weis,
1985). Since students’ experiences in the broader community may be less structured, less
supervised and supported, and less timetabled, the community context stands in contrast
to school as students make decisions and find opportunities to participate in drug userelated activities. This draws attention to the importance of developing community-wide
approaches for encouraging pro-social behaviour.

Implications for Schools
There are also practical implications. Schools, as primary socializing institutions,
play a central role in the development of youth (Antro-Gonzalez & De Jesus, 2006;
Hawkins, Catalano & Catalano, 1995; Hirschi, 1969; cited in Moyer, 2001). The notion

90

of school bonding emphasizes the need for schools to develop a systemic prevention
strategy built on a pro-active model of health, support, and engagement (Hawkins,
Catalano & Catalano, 1995). This stands in contrast to current models in use that are
largely oriented toward identifying and reacting to the ‘unruly’, delinquent, or
problematic student (McCrystal, Percy & Higgins, 2007).
Developing a more ‘connected’ school is a modifiable factor (Thomas & Smith,
2004). When the philosophical orientation of a school is based on the deficit model, and
when alienating and exclusionary practices, such as school suspensions or other punitive
forms of discipline, become the dominant response to delinquency, social programs of a
preventative nature are more likely to fail (Hawkins, Arthur & Catalano, 1995).
Braithwaite (1999) reiterates the importance of adopting relational philosophies in
schools, based on a wide body of evidence that is showing student engagement and
school belonging to be an effective and proactive prevention strategy (cited in McCrystal,
Percy & Higgins, 2007). Increasingly, school exclusion is being viewed as ‘bad policy’
because it abandons youth who are already at higher risk for using drugs, and among
those who are at increased risk for long term problems (Lewis, Sugai & Colvin, 1998;
McCrystal, Percy & Higgins, 2007). Shifting a disproportionate level of responsibility to
the individual in the absence of social support suggests that drug use is largely a matter of
individual choice and freewill. This ultimately minimizes the responsibility that schools
and government have as “duty bearers” to support families in meeting the task of raising
children to reach their full potential (Department of Canadian Heritage, 1991).
From this study, it became evident that school attachment indicators provided a
perceptual indicator of risk; while school involvement indicators marked specific

91

behaviours for mitigating risk as a matter of daily prevention practice. This study
confirms current research identifying that students who report a ‘connected education’those who found their school environment to be socially supportive, and who liked school
- were less likely to deviate from pro social norms (Zyngier, 2003). Antrop-Gonzales and
DeJesus (2007) draw attention to the reciprocal and mutually beneficial nature of secure
relational attachments at school. This is influenced by the degree to which students
experience high quality interactions with school staff by observing their teachers support
them through fair and respectful processes, and having teachers empower students in
ways that help them attain educational standards in a consistent fashion (Butler, Leschied
& Fearon 2007; Klen & Connell, 2004). Schools can alleviate risk for drug use by
providing learning opportunities that deliver a sense of meaning, purpose, and sufficient
challenge to students’ lives (Larson, 2001). The literature also tells us that this is most
likely to occur in schools that can provide an enjoyable learning environment, where
relationships between students and teachers are supportive, where school staff and
students participate jointly in school activities, and where students are given opportunities
to develop skills through leadership roles (Henry & Slater, 2007). When students do not
feel secure, included, and supported toward meeting educational standards, their
motivation to engage in pro-social behaviour is undermined (Goodenow, 1992; Klem &
Connel, 2004).
This research brought attention to specific behavioural markers, such as skipping
class, and skipping an entire day of school, as an aspect of school bonding. These
variables not only provided the strongest predictor among those examined for current and
eventual drug use behaviour, but also delineated behavioural markers for including in risk

92

assessment strategies. Previous research has linked school absenteeism to a variety of risk
behaviours including drug and alcohol use, violence, risky sexual activity, and
unintentional injury (Eaton, Brener & Kann, 2008). This was the case even under
conditions where students had received prior permission from adults to be absent. For
reasons beyond preventing substance use alone, it is critical for schools to recognize
absenteeism as a potential marker for intervention due to its predictable association to
drug use. Eaton, Brener & Kann’s study also draw links between absenteeism and
physical and mental health issues, negative school environments, dysfunctional home
environments, poverty, and low academic motivation. The most salient efforts to improve
student attendance occurred through coordinated and comprehensive intervention designs
that drew involvement from students, parents, educators, and the broader community
(Epstein & Sheldon, 2002; cited in Eaton, Brener & Kann, 2008; Hawkins, Arthur &
Catalano, 1995).
Schools are in a prime position to coordinate support through their centralized
location (Eaton, Brener & Kann, 2008; Hawkins, Arthur & Catalano, 1995). The Social
Development Model presents a health framework for strengthening the protective and
functional role of schools (Hawkins & Weis, 1985; Henry & Slater, 2007). Experts
(Hawkins & Weis, 1985) maintain the most significant targets for intervention are at the
family, the school, and peer level, depending on the developmental stage of the child.
Accordingly, in succession, families are the most appropriate from early childhood to
early adolescence; schools are the most suitable throughout the educational years, and
peers become particularly relevant during adolescence. Within each unit, positive youth
development is said to result when youth are given the opportunity to become involved in

93

conforming activities, as they are able to develop the skills and attitudes necessary for
successful involvement in those activities, and when they are recognized for behaving in
socially approved ways. The resulting growth of social bonds involving attachment to
pro-social others not only engenders belief in a conventional order, but also increases
commitment to conforming behaviour, while decreases the likelihood of developing
associations to delinquent peers, and eventual delinquency (Hawkins & Weis, 1985).
The prevalence of drug use among youth holds significant challenge for schools.
Although youth who were more socially bonded to school in this study were less likely to
use drugs, the research states that drug use has made its way into mainstream youth
culture to the degree that its use is no longer limited to delinquent or disenfranchised
youth (CCSA, 2007b; McCrystal, Percy & Higgins, 2007). Students may be able to
comply with the norms while in school as a way to get through the day, however,
opportunities for using drugs in the community may provide an escape from a general
dissatisfaction related to boredom levels at school (Larson, 2000). Being surprised by the
sheer numbers of students who reported a high degree of dissatisfaction with their school
experience, Larson captured the sentiment among sufficient numbers of students who
claim they are “being trapped in the present, waiting for someone to prove to them that
life is worth living” (2000:170). This suggests deficiencies in positive youth development
(Larson, 2000; Hawkins, Arthur & Catalano, 1995).
Further research is needed to understand how school bonding interacts with
competing external pressures, including the extent to which drug use among mainstream
youth exists. It appears as though for at least a proportion of the student population,
substance use may be a way of enhancing mood, dealing with dissatisfaction, asserting

94

independence, or bringing a sense of pleasure to life even among those from the
mainstream (CCSA, 2007b). While adolescence is acknowledged to be a time of risktaking, recent literature acknowledges the normalisation of drugs as “reflect[ing]
the...emergence of a new, post-modern, social order in which young people are, without
doubt, the prime users of illicit drugs” (Shiner & Newburn, 1999; cited in McCrystal,
Percy & Higgins 2007:38). Taken together, the growing normalisation of drugs, along
with student dissatisfaction about their educational experiences, sets up a climate of risk
that can be observed even among those that society might least expect. This suggests that
primary youth serving structures, such as schools, should continue to evolve policies in a
way that engenders school bonding as a way to provide resiliency against the use of
drugs.

Limitations
Although this study offered new insights, several limitations are important to
recognize. The first limitation concerns sample characteristics and the extent to which the
findings can be generalized to the broader population of youth. Data was collected
through a blanket survey process at school for those attending and willing to complete the
questionnaire. Accordingly, it fails to account for street youth, those who have dropped
out of school, youth on probation, and non-English speaking students. In addition, of the
2,477 secondary students from both urban and rural schools, the sample used in this
analysis isolated those who chose to complete the privacy code in the same manner for
both years. Arguably then, the sample is skewed in favour of students who may be more
bonded to school from the outset for various reasons. This may have implications for

95

assessing the relationship between school bonding and drug use given that survey
participation is form of ‘school involvement’ itself. However, this may also suggest that
the findings in this study are conservative in demonstrating the magnitude of the
relationship between school bonding and drug use.
An additional limitation worth noting is that this study has only considered school
bonding components using three measures each from the school attachment domain
(school liking, respect from adults at school, and recognition from a school adult
concerning the students’ ideas and opinions), and the school involvement domain
(skipping class, skipping a day, and participating in school groups of clubs). This
limitation is due to the use of using secondary data to answer the research questions. As
such, further research would be required in order to conduct a more comprehensive
examination of the school bonding notion using elements from Social Bonding Theory.

Conclusion
This study provided evidence in support of the proposition that school bonding,
examining school attachment and school involvement, is a predictor of both current and
future drug use. It would appear as though healthy relationships are at the foundation of
pro-social individuals and communities, while participation in school activities provides a
venue for developing, maintaining, and strengthening those ties. While a significant
proportion of youth across grades eight to eleven reported the use of drugs, the focus of
this study was to gain a better understanding of school-related factors that could mitigate
drug use and increase our understanding for the role of schools in facilitating the wellbeing of youth.

96

Though this study was unable to determine causality according to the three
specific criteria outlined by Paul Lazarsfeld (1959; cited in Babbie, 2001), it has been
able to establish that a statistically significant inverse relationship exists between at least
two components of Hirschi’s Social Bonding Theory and drug use. It further determined
that this relationship also holds true within a temporal analytic research design in which
school bonding became a preceding condition to drug use among a sub-sample of school
youth. While the relationship between these two variables was weaker using an across
time analysis, it nonetheless suggested that schools, while not alone in their responsibility
to address drug use, have an important role to play in preventing current and future drug
use.
Society’s investment in promoting high quality educational environments
becomes apparent when noting the far-reaching costs of drug use (CCSAb, 2007; Health
Canada, 2001). This suggests that schools should be encouraged to, and supported in,
making further investments to improve processes, and enhance programs that, promote
positive engagement and community building in schools (Payne, Gottfredson &
Gottfredson, 2003). Research should also continue to identify the particular mechanisms
and programs that can assist schools for enhancing their capacity to facilitate positive
youth development, in particular, for those most at risk. Experiencing social and
educational support is already noted for its ability to contribute to a more comprehensive
scaffolding to enable all youth to develop within pro-social climates of care (Bonny et al.
2000).
The research findings also suggest that rule-breaking behaviour, such as skipping
school, may be more effectively viewed within a risk perspective for which protective

97

intervention strategies form a best response (Braithwaite, 1999; Hawkins, Arthur &
Catalano, 1995). It would appear that interventions for disenfranchised youth can realize
greater gains by strengthening the psychological connections that youth have with adults
at school. While school curriculum and programming appear to reduce risk for drug use
at school, it is also important to implement community models that embrace youth
engagement and facilitate social support in response to drug use in the community.
Schooling continues to be one of the most important investments that society can offer its
youth for a better future, while drug use may well symbolize the destruction of that goal.

98

References

Adlaf, E., & Racine, S. (2005). Alcohol and other drugs. In Health Canada (Ed.), 2002
Youth Smoking Survey Technical Report. Ottawa: Ministry of Supply and Services
Canada. Retrieved January 9, 2008, from http://www.hc-sc.gc.ca/hlvs/alt_formats/hecs-sesc/pdf/pubs/tobac-tabac/yss-etj-2002/chap-11-eng.pdf.

Akers, R. & Lee, G. (1999). Age, social learning and social bonding in adolescent
substance use. Retrieved October 15, 2007, from
http://digitalcommons.utep.edu/cgi/viewcontent.cgi?article=1005&context=gang_lee

Anderman, L. (2003). Academic and social perceptions as predictors of change in middle
school students' sense of school belonging. Journal of Experimental Education,
72(1), 5-22. Retrieved on May 5, 2008 from ERIC database.

Antrop-Gonzalez, R. & De Jesus, A. (2006). Toward a theory of critical care in urban
small school reform: examining structures and pedagogies of caring in two Latino
community-based schools. International Journal of Qualitative Studies in Education.
19(4), 409-433. [Electronic version]. Retrieved November 15, 2007, from ERIC
database.
Babbie, E. (2001). The Practice of Social Research. 9th ed. Toronto: Nelson Thomson
Learning.

99

Bonny, A., Britto, M., Klostermann, B., Hornung, R., & Slap, G. (2000). School
disconnectedness: Identifying adolescents at risk. Pediatrics, 106. Retrieved January
5, 2007 from http://www.pediatrics.org/cgi/content/full/106/5/1017

Butler, S., Leschied, A., & Fearon, P. (2007). Antisocial beliefs and attitudes in preadolescent and adolescent youth: The developing of the antisocial beliefs and
attitudes scales (ABAS). Journal of Youth Adolescence. 36(8), 1058-1071.

Canadian Centre on Substance Abuse. (2007a). A drug prevention strategy for Canada’s
youth. Retrieved March 30, 2008, from
http://www.ccsa.ca/NR/rdonlyres/EC53A270-E4C5-46B0-9650AD137DB63E9F/0/ccsa0115222007e.pdf

Canadian Centre on Substance Abuse. (2007b). Substance abuse in Canada: Youth in
focus. Ottawa, ON: Canadian Centre on Substance Abuse. Retrieved March 30,
2008, from http://www.ccsa.ca/NR/rdonlyres/5D418288-5147-4CAC-A6E46D09EC6CBE13/0/ccsa0115212007e.pdf

Canadian Centre on Substance Abuse. (2007c). A constant presence in a changing field.
2006-2007 Annual Report, Ottawa, ON. 2006-2007 Annual Report. Ottawa, ON.

Canadian Health Network. (2007). Growing up, taking risks. Public Health Agency of
Canada. Retrieved December 12, 2007, from http://www.canadian-healthnetwork.ca/servlet/ContentServer?cid=1172770076956&pagename=CHNRCS/CHNResource/CHNResourcePageTemplate&c=CHNResource

100

Centre for Addiction and Mental Health. (2008). Defining addiction, dependence and
abuse. Retrieved October 1, 2008 at
http://www.camh.net/Publications/Resources_for_Professionals/Pregnancy_Lactatio
n/defining_addiction.html.

Cohen, I., Plecas, D., & Watkinson, A. (2005). A review of the research on the Drug
Abuse Resistance Education (D.A.R.E.) program. Abbotsford: University of the
Fraser Valley, School of Criminology and Criminal Justice.

Collier, M. (2006). A structure for caring in schools. Journal of Human Behaviour in the
Social Environment, 13(4), 73-83.

Currie, C., Roberts, C., Morgan, A., Smith, R., Settertobulte, W., Samdal, O., Barnekow,
R. (2004). Young people’s health in context: Health behaviour in school-aged
children (HBSC). International Report from the 2001/2002 Survey: World Health
Organization; 2004. Health Police for Children and Adolescents, No. 4. [Electronic
version]. Retrieved October 28, 2007, from
http://www.euro.who.int/eprise/main/who/informationsources/publications/catalogue
/20040518_1 .

101

DeWit, D., Akst, L., Braun, K., Jelley, J., Lefebvre, L., McKee, C., Rye, B., & Shain, M.
(2002). Sense of school membership: A mediating mechanism linking student
perceptions of school culture with academic and behavioural functioning. Centre for
Addiction and Mental Health. Retrieved February 2007 from
http://www.camh.net/About_Addiction_Mental_Health/Child_Youth_Family_Resou
rces/school_culture_report.pdf .

Department of Canadian Heritage (1991). Convention on the Rights of the Child . Human
Rights Program Publication No. S2-210/1991E). Hull QC: Minister of Supply and
Service Canada.

Diaz, J. (2005). School attachment among Latino youth in rural Minnesota. Hispanic
Journal of Behavioural Sciences, 27(3), 300-318.

Eaton, D., Brener, N., & Kann, L. (2008). Associations of health risk behaviours with
school absenteeism. Does having permission for the absence make a difference?
Journal of School Health, 78(4), 223-229.

Eith, C. (2005). Students, schools, and the social bond: An analysis of school bonding
from elementary to high school. Dissertation Abstracts International, 66(5), 1964-A.

Enomoto, E. (1997). Schools as nested communities. Urban Education, 32(4), 512-531.

Faircloth, B., & Hamm, J. (2005). Sense of belonging among high school students
representing 4 ethnic groups. Journal of Youth and Adolescence, 34(4), 293-309.

102

Frey, L., Botan, C., & Kreps, G. (1999). Investigating communication: An introduction to
research methods. NJ: Allyn & Bacon.

Gandhi, A., Murphy-Graham, E., Petrosino, A., Chrismer, S., & Weiss, C. (2007). The
Devil is in the details: Examining the evidence for “proven” drug abuse prevention
programs. Evaluation Review, 31(1), 43-74. Retrieved September 15, 2007 from
Sage Publication database.

Garcia-Reid, P., Reid, R., & Peterson, N. (2005). School engagement among Latino
youth in an urban middle school context. Education and Urban Society, 37(3), 257275.

Goodenow, C. (1992). School motivation, engagement, and sense of belonging among
urban adolescent students. Paper presented at the Annual Meeting of the American
Educational Research Association, San Fransisco, CA. 1-30. Retrieved April 2,
2007, from
http://eric.ed.gov/ERICDocs/data/ericdocs2/content_storage_01/0000000b/80/22/5a/
2c.pdf

Gottfredson, M., & Hirschi, T. (1990). A General Theory of Crime. Stanford, CA:
Stanford University Press.

Gravetter, F., & Wallnau, L. (2002). Essentials of Statistics for the Behavioural Sciences.
(4th ed.). Pacific Grove, CA: Wadsworth.

103

Hall, W., Doran, C., Degenhardt, L. & Shepard, D. (2006). “Illicit Opiate Abuse”. In
Disease Control Priorities in Developing Countries (2nd ed.). 907-932. New York:
Oxford University Press. Retrieved June 16, 2007 from
http://www.dcp2.org/pubs/DCP/48/FullText

Hawkins, J., & Weis, J. (1985). The social development model: An integrated approach
to delinquency prevention. Journal of Primary Prevention, 6(2), 73-97.

Hawkins, J., Arthur, M., & Catalano, R. (1995). Preventing substance use. Crime and
Justice. Building a Safer Society: Strategic Approaches to Crime Prevention.
Retrieved December 1, 2007 from http://www.jstor.org/cgibin/jstor/printpage/01923234/ap040019/04a00080/0.pdf?backcontext=page&dow
hat=Acrobat&config=jstor&userID=c6a2600d@ucfv.ca/01c0a80a6800501d1fe81
&0.pdf
Health Canada (2001). Preventing substance use problems among young people: A
compendium of best practices. Canadian Centre on Substance Abuse for the
Office of Canada’s Drug Strategy. Retrieved on December 1, 2007 from
http://www.hc-sc.gc.ca/hl-vs/alt_formats/hecs-sesc/pdf/pubs/adpapd/prevent/young-jeune_e.pdf

Henry, K. L., & Slater, M. D. (2007). The contextual effect of school attachment on
young adolescents' alcohol use. Journal of School Health, 77(2), 67-74.

Hirschi, T. (1969). Causes of Delinquency. New Brunswick, NJ: Transaction Publishers.

104

Hoeve, M., Smeenk, W., Loeber, R., Stouthamer-Loeber, M., van der Laan,Peter H.,
Gerris, J. and Dubas, J. (2007). Long-term effects of parenting and family
characteristics on delinquency of male young adults. European Journal of
Criminology, 4(2) 161-194. Retrieved on January 16, 2008 from Sage Publication
Database.

Hoppe, M., Wells, E., Haggerty, K., Simpson, E., Gainey, R., & Catalano, R. (1998).
Bonding in a high-risk and a general sample of children: Comparison of measures of
attachment and their relationship to smoking and drinking. Journal of Youth and
Adolescence, 47(1), 59-81.

Hotton, T., & Haans, D. (2004). Alcohol and Drug Use in Early Adolescence. Health
Reports, 15(3). Statistics Canada, Catalogue 82-003. Retrieved September 12, 2007
from http://www.statcan.ca/english/studies/82-003/archive/2004/15-3-a.pdf.

Hunt, G., Evans, K., & Kares, F. (2007). Drug use and meanings of risk and pleasure.
Journal of Youth Studies, 10(1), 73-96.

Hwang, S., & Akers, R. (2003). Substance use by Korean adolescents: A cross-cultural
test of social learning, social bonding, and self-control theories. In R. Akers & G.
Jensen (Eds.). Social Learning Theory and the explanation of crime. (Vol. 11). P.
39-63. Rutgers, New Jersey:Transaction.

Jarvinen, M., & Gundelach, P. (2007). Teenage drinking, symbolic capital and
distinction. Journal of Youth Studies, 10(1), 55-71.

105

Johnson, L., O’Malley, P., Bachman, J., & Schulenberg, J. (2006). Monitoring the Future
national survey results on drug use: Overview of key findings 2006. (NIH
Publication No. 07-6202).

Kane, J. (2006). School exclusions and masculine, working-class identities. Gender and
Education, 18(6), 673-685.

Katz, M., Noddings, N., & Strike, K. (1999). Justice and caring: The search for common
ground in education. New York: Teachers College Press.

Klem, A., & Connell, J. (2004). Relationships matter: Linking teacher support to student
engagement and achievement. Journal of School Health, 74(7) 1-20.

Kohen, D., Dahinten, V., Leventhal, T., & McIntosh, C. (2008). Neighborhood
disadvantage: pathways of effects for young children. Child Development, 79(1), p.
156-169.

Krohn, M., Lizotte, A., & Perez, C. (1997). The interrelationship between substance use
and precocious transitions to adult statuses. Journal of Health and Social Behaviour
38(1), 87-103.

LaGrange, H., Pandina, R. and LaGrange, W. (2006). Longitudinal predictors of serious
substance use and delinquency. Criminology, 25(3), p. 715-740. Retrieved February
19, 2008 from Wiley Interscience at http://www3.interscience.wiley.com/journal

Larson, R. (2000). Toward a psychology of positive youth development. American
Psychologist, 55(1), 170-183.
106

LaRusso, M., Romer, D. & Selman, R. (2007). The roles and perspectives of school
mental health professions in promoting adolescent mental health. In D.L. Evans,
E.B. Foa, R.E. Gur, H. Hendin, C.P. O’Brian, M. E.P.Selignan, & B.T. Walsh (Eds.),
Treating and preventing adolescent mental health disorders (pp. 579-596). New
York: Oxford University Press. Retrieved February 19, 2008 from http://amhitreatingpreventing.oup.com/anbrg/public/content/mentalhealth/9780195173642/toc.h
tml

Lewis, T., Sugai, G., & Colvin, G. (1998). Reducing problem behaviour through a
school-wide system of effective behavioural support: Investigation of a school-wide
social skills training program and contextual interventions. School Psychology
Review, 27(3), p.446-459.

McCrystal, P., Percy, A., & Higgins, K. (2007). Exclusion and marginalization in
adolescence: The experience of school exclusion on drug use and antisocial
behaviour. Journal of Youth Studies 10(1), 35-54.
McLaughlin E. & Muncie, J. (Ed.). (2006). The SAGE dictionary of criminology, (2nd
Ed.). Thousand Oaks: Sage.

Ma, X. (2003). Sense of belonging to school: Can schools make a difference? Journal of
Educational Research, 96(6), p. 340-349.

Maddox, S., & Prinz, R. (2003). School bonding in children and adolescents:
Conceptualization, assessment, and associated variables. Clinical Child and
Family Psychology Review, 6(1), 31-49.
107

Marsiglia, F., Miles, B., & Dustman, P. (2003). Ties that protect: An ecological
perspective on Latino/a urban pre-adolescent drug use. Journal of Ethnic and
Cultural Diversity in Social Work, 11(3/4), 191-220.

Mosher, C., Rotolo, T., Phillips, D., Krupski, A., & Stark, K. (2004). Minority
adolescents and substance use risk/protective factors: A focus on inhalant use.
Adolescence, 39(155), 489-502.

Moyer, I. (2001). Criminological theories: Traditional and nontraditional voices and
themes. Thousand Oaks: Sage Publications.

Murray, C., & Greenberg, M. (2000). Children’s relationship with teachers and bonds
with school: an investigation of patterns and correlates in middle childhood. Journal
of School Psychology, 38(5), 423-445.

Napoli, M., Marsiglia, F., & Kulis, S. (2003). Sense of belonging in school as a protective
factor against drug abuse among Native American urban adolescents. Journal of
Social Work Practice in the Addictions, 3(2), 25-41.

Noddings, N. (1984). Caring: A feminine approach to ethics and moral education.
Berkeley: University of California Press.

Noddings, N. (1992). The challenge to care in schools: An alternative approach to
education. New York, NY: Teachers College Press.

108

Pardini, D., Loeber, R., & Stouthamer-Loeber, M. (2005). Developmental shifts in parent
and peer influences on boys’ belief about delinquent behaviour. Journal of Research
on Adolescence, 15(3), 299-323.

Payne, A., Gottfredson, D., & Gottfredson, G. (2003). Schools as Communities: The
relationships among communal school organization, student bonding, and school
disorder. Criminology, 41(3), 749-777.

Robertson, E., David, S., & Rao, S. (2003). Preventing drug use among children and
adolescents: A research-based guide for parents, educators, and community leaders.
(2nd ed.). Bethesda, Maryland: U.S. Department of Health and Human Services.
Downloaded December 2, 2007 from website
http://www.drugabuse.gov/pdf/prevention/InBrief.pdf

Scales, P. (2005). Developmental assets and the middle school counselor. Professional
School Counselling, 9(2), p. 104-111.

Schaps, E., & Solomon, D. (2003). "The role of the school's social environment in
preventing student drug use. Journal of Primary Prevention, 23(3), 299-328.

Sergiovanni, T. (1994). Organizations of communities? Changing the metaphor changes
the theory. Educational Administration Quarterly, 30(2), 214-226.

Siegel, L., & McCormick, C. (1999). Criminology in Canada: Theories, patterns, and
typologies. Toronto, ON: International Thomson.

109

Simons-Morton, B., Crump, A., Haynie, D., & Saylor, K. (1999). Students-school
bonding and adolescent problem behaviour. Health Education Research, 14(1),
99-107.

Substance Abuse and Mental Health Services Administration (2002). Results from the
2001 National Household Survey on Drug Abuse: Volume 1, Summary of National
Findings (NHSDA Series H-17, DHHS Publication No. SMA 02-3758). Rockville,
Maryland: Office of Applied Studies.

Sweet, S., & Grace-Martin, K. (2003). Data Analysis with SPSS: A First Course in
Applied Statistics. Toronto, ON: Pearson Education.

Thomas, S., & Smith, H. (2004). School connectedness, anger behaviours, and
relationships of violent and nonviolent American youth. Perspectives in Psychiatric
Care, 40(4), 135-148.

United Nations. (2005). World Youth Report: Young people today and in 2015.
Department of Economic and Social Affairs. ISBN 92-1-130244-7. Retrieved
October 1, 2007, from
http://www.un.org/esa/socdev/unyin/documents/wyr05book.pdf

U.S. Department of Health and Human Services. (2005). National Survey on Drug Use &
Health: National Results. Retrieved July 10, 2007, from
http://www.oas.samhsa.gov/drugs.cfm

110

Vancouver Coastal Health News Release (2006). This is not a test: 2006 Vancouver
Youth Drug Survey Results. Retrieved on March 15, 2008 from
http://www.vch.ca/news/docs/2007_03_06_YouthDrugSurveyReport.pdf

Voelkl, K. & Frone, J. (2001, May). Predictors of substance use at school among high
school students. Research In Brief, Article, 1, 1-4.

Weiner, D., Abraham, M., & Lyons, J. (2001). Clinical characteristics of youths with
substance use problems and implications for residential treatment. American
Psychiatric Association, Psychiatric Services. 52, 793-799.

Wotherspoon, T. (2004). The Sociology of Education in Canada. Don Mills, On:Oxford.
Zyngier, D. (2003). Connectedness – Isn’t it time that education came out from behind
the classroom door and rediscovered social justice. Social Alternatives, 22(3), 41-49.

111

APPENDIX A – SAFE SCHOOL SOCIAL RESPONSIBILITY
SURVEY FOR SECONDARY STUDENTS

112

113

114

115

116

117

118

119

120

121

122

123

124

125

126

127

128

129

130