The Importance of School

• Author: Anthony N. Doob,Jennifer M. Jenkins,Jane B. Sprott
• Published date: 01 January 2005
• Date: 01 January 2005
• DOI: http://doi.org/10.1177/1541204004270943
• Subject Matter: Articles

10.1177/1541204004270943
Youth Violence and Juvenile Justice
Sprott et al. / DELINQUENCY AND SCHOOL BOND
THE IMPORTANCE OF SCHOOL
Protecting At-Risk Youth From Early Offending
Jane B. Sprott
University of Guelph
Jennifer M. Jenkins
Anthony N. Doob
University of Toronto
Using longitudinal data, this study explored the protective effect of a school bond (on
violent and nonviolent delinquency) across a variety of risk factors. A strong school
bond protected children with early aggression from continuing on in violence 2 years
later. A strong school bond also protected children with numerous environmental risks
from violent and nonviolent offending. Finally, a strong school bond acted as a protec-
tive factor against the influence of peers who were delinquent (for nonviolent offending
only). From a policy perspective, these findings suggest that zero-tolerance polices that
aim to exclude children considered to be a problem through suspensions or expulsions
could be counterproductive.
Keywords: risk and protective factors; preventing delinquency; school bond; school
attachment
Although the study of risk and resiliency has been a field of inquiry for developmen-
tal psychologists for some time, criminologists have only relatively recently adopted such a
model to understand the development of offending behavior. Most insight in criminological
research has been gained on identifying and understanding risk factors related to offending
(Farrington, 2000). Generally, delinquent behavior is thought to be the result of complex
interactions of individual traits with social (e.g., family, peers), situational (e.g., school,
home), and neighborhood risk factors (Farrington, 1998; Gottfredson & Hirschi, 1990;
Klein, Forehand, Armistead, & Long, 1997; Loeber & Farrington, 1998; Sampson & Laub,
1993). What is less well understood in criminological research is why some individuals
from high-risk environments do not become seriously involved in delinquency. The phe-
nomenon of remaining healthy in spite of serious adversity has been labeled resilience. In
investigating resiliency, researchers have sought to identify protective factors that reduce
levels of impairment in high-risk situations.
Authors’ Note: The preparation of this article was financially supported by the Applied Research Branch of
Human Resources Development Canada and by research grants from the Social Sciences and Humanities Research
Council of Canada to each of the three coauthors.
Youth Violence and Juvenile Justice, Vol. 3 No. 1, January 2005 59-77
DOI: 10.1177/1541204004270943
© 2005 Sage Publications
59

---

60 Youth Violence and Juvenile Justice
Risk and Protective Factor Paradigm
There has been some confusion over the difference between risk factors and protec-
tive factors. Conceptually, protective factors could be seen as simply the opposite of risk
factors. For example, family environments that are dysfunctional increase the risk of chil-
dren developing various problems whereas healthy family environments decrease the risk
of children developing problems. Thus, some have argued that for the study of resiliency to
provide insight over and above that gained by the study of risk, we have to look for interac-
tions between risk and protective factors (Rutter, 1983); that is, we gain new insight into
how children in high-risk situations cope when we can isolate a factor that appears to
decrease the level of disturbance in children who are high risk but has little or no effect on
children who are lower risk.
For the purpose of this article, a factor that produces only a main effect will be called a
risk factor, and the term protective is reserved for those factors that produce statistical inter-
actions between individuals who are high and low risk. Two types of protective effects rele-
vant to the current study have been described in the literature. Some protective factors are
not associated with decreased disturbance in the low-risk group but show a large effect in
the high-risk group. Other protective effects have been described in which the protective
factor is significantly associated with decreased disturbance in the high- and the low-risk
groups; however, the association in the high-risk group is much stronger. In this circum-
stance, the protective factor plays much more of a role in lowering disturbance in the high-
risk group but also matters to development in the low-risk group. Cohen and Wills (1985)
called this “protective at high and low risk.”
The School Bond as Protection
The idea that school plays an important role in children’s lives is not new. In a 1936
study by Healy and Bronner, it was found that “about 40% of the 105 delinquent children
expressed marked dislike for school in general . . . and . . . a mere 4% of the controls [a
nondelinquent sibling] evinced any such dislikes” (p. 62). Similarly, in that three-city com-
parison of siblings who were delinquent and were nondelinquent, it was found that “defi-
nitely poor scholarship was registered for . . . 34% of the delinquents as against 18% of the
controls” (p. 61). For almost 70 years then, there has been little dispute that school is associ-
ated with children’s delinquency.
In more recent research, school has consistently been identified as a risk factor for
delinquency (see, e.g., Agnew & Brezina, 1997; Blumstein, Farrington, & Moitra, 1985;
Elliot, Huizinga, & Menard, 1989; Krohn & Massey, 1980; Sampson & Laub, 1993;
Thornberry, Lizotte, Krohn, Farnworth, & Jang, 1991; Thornberry, Moore, & Christenson,
1985; J. Williams, Ayer, Abbott, Hawkins, & Catalano, 1999; Zingraff, Leiter, Johnson, &
Myers, 1994); that is, weak school bonds increase the likelihood that children will engage in
delinquency. Generally, “Children who do poorly in school, reject the authority of schools
and their officials, have low aspirations concerning their probable life chances upon com-
pletion of school, and drop out of school before graduation are at risk of subsequent delin-
quency” (Zingraff et al., 1994, p. 67).
Hirschi’s (1969) social control theory is the theory most widely used to explain the
school-delinquency relationship. Social control theory suggests that the school and school
experiences serve as social bonds that restrain children and adolescent involvement in
delinquency. Hirschi identified four elements of the social bond: attachment (caring about

---

Sprott et al. / DELINQUENCY AND SCHOOL BOND 61
others and what others think), commitment (commitment to educational values), involve-
ment (participating in school-related activities), and belief (accepting school rules, and
school authority as fair). Hirschi hypothesized that these elements of the social bond work
to build a stake in conformity and thus limit involvement in normatively unconventional
activities.
Many researchers have investigated the relative importance of each of the four ele-
ments of the school bond. Several have found that poor educational achievement and aca-
demic failure are most strongly and consistently linked with delinquency (see, e.g.,
Gottfredson & Hirschi, 1990; Hirschi, 1969; P. Jenkins, 1997; Loeber & Dishion, 1983;
Maguin & Loeber, 1996; Zingraff et al., 1994). However, although school commitment
and grades have been identified as risk factors for involvement in delinquency, a question
remains about whether the school bond is simply a risk factor or whether elements of the
school bond might have an additional function—that of a protective factor.
There is some evidence that various elements of the school bond protect children who
are high risk. Tiet et al. (1998), for example, explored factors that characterized the resilient
children from high-risk environments. They found that
children at high risk [for numerous problem behaviors] because of higher levels of
adverse life events exhibited a greater degree of resilience when they had a higher IQ,
better family functioning, closer parental monitoring, more adults in the household, and
higher educational aspirations. (Tiet et al., 1998, p. 1191)
Thus a strong commitment to school and education served to protect children who were
high risk. This suggests that school bonds may act as protective factors. In this article, we
investigate the protective effect of school across three different risks: early aggression, a
cumulative environmental risk index, and a delinquent peer group. Specifically, we used
two cycles of Canadian longitudinal data to examine the protective effect of a strong school
bond (assessed at age 10 and 11 years) on early involvement in violent and nonviolent
offending (assessed at age 12 and 13 years) across our three different risks (assessed at age
10 and 11 years).
In all analyses, violent and nonviolent offending were kept separate from one
another. This was done because the protective effect of a strong school bond may affect
these two types of offending in different ways. In addition, children are much more likely to
be involved in property offending than violent offending at this relatively young age. Given
the differential involvement in violent versus nonviolent offending, it may be the school
bond and risk interactions produce different results depending on the type of delinquency
investigated.
Risks
Based on the data available, we chose to examine three theoretically relevant risks:
early aggression, cumulative risks, and peer deviance. The high level of continuity in chil-
dren’s aggression means that previous history of aggression is...