The development of early delinquency: can classroom and school climates make a difference?

• Author: Sprott, Jane B.

Previous research has found that school and classroom climates have important effects on children's perceptions and behaviours. More specifically, there are thought to be two types of support (emotional and instrumental) provided at the level of the classroom and the school. Emotional support within the classroom has been found to be most important for some higher-risk children. There has, however, been little research using these concepts with outcomes such as delinquency. Therefore, using two years of the Canadian National Longitudinal Survey of Children and Youth, this study investigates the role of classroom and school climates on the development of early violence and property offending. Results revealed that an emotionally supportive classroom when these children were 10 to 13 years old was related to lower levels of violence two years later, when they were 12 to 15 years old. In addition, a classroom that focused on academics (instrumental support) was predictive of lower property offending. Interpretations and policy implications are discussed.

Certaines recherches ont mis en lumiere les effets importants de l'ecole et de la salle de classe sur les perceptions et les comportements des enfants. Plus precisement, c'est dans ces lieux que les eleves beneficieraient a la fois d'un soutien affectif et d'un soutien instrumental. Ainsi, on constate que le la salle de classe apporte un soutien affectif particulierement important a certains enfants a risque eleve. Cependant, on a effectual tres peu de recherches sur l'application de ces types de soutien des comportements tels que la ddlinquance. Dans la presente etude, l'auteure essaie donc de rectifier cette lacune en examinant, a partir de donndes portant sur deux annees de l'Enquete longitudinale nationale sur les enfants et les jeunes, les effets de l'environnement etabli dans la salle de classe ou dans l'ecole sur l'dmergence prdcoce de la violence et des infractions contre les biens. Or, les rdsultats de l'etude demontrent un lien entre le soutien affectif important offert en salle de classe a des enfants de 10 a 13 ans et une baisse du niveau de violence, deux ans plus tard, chez ces memes enfants a l'age de 12 a 15 ans. Par ailleurs, l'analyse des donnees permet de predire que la salle de classe axee sur l'enseignement des matieres scolaires (soutien instrumental) susciterait une baisse du nombre d'infractions contre les biens. L'auteure discute egalement de l'interpretation des resultats et de leurs consequences pour l'elaboration des politiques dans ce domaine.

Introduction

There is little debate that schools play an important role in children's lives. Most criminological research draws upon Hirschi's (1969) social control theory to understand how schools relate to delinquency. Specifically, social control theory suggests that the school and school experiences serve as social bonds that restrain child and adolescent involvement in delinquency. Hirschi identified four elements of the social bond: attachment (caring about others and what others think); commitment (commitment to educational values); involvement (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. Research has generally supported Hirschi's theory (see, e.g., Jenkins 1995, 1997; Maguin and Loeber 1996; Zingraff, Leiter, Johnson, and Myers 1994). Youths who are attached and committed to school, are highly involved in school, and have positive beliefs or aspirations towards school and school achievement are much less likely to become seriously involved in delinquency and drug use (see Gottfredson 2001 for a more detailed review).

While social control theory focuses on the school bond, other research has examined aspects of the school climate that affect behaviours. One study assessed the independent effects of individual and school characteristics on a variety of outcomes (Rutter, Maughan, Mortimore, and Ouston 1979). The results showed that when individual factors were controlled, school characteristics (e.g., rewards and punishments; good working conditions; responsiveness to students' needs; clear academic goals; good group management in the classroom; firm leadership; a balance of intellectually able and less able students) were all related to positive school outcomes (e.g., academic achievement, regular attendance). Later studies have generally resulted in similar findings (see, e.g., Mortimore, Sammons, Stoll, Lewis, and Ecob 1988; Caspi, Lynam, Moffitt, and Silva 1993; Kasen, Johnson, and Cohen 1990; or, for a review of the research, see Gottfredson 2001).

Another landmark study, conducted by Gottfredson and Gottfredson (1985), involved school data from a 1976 sample of more than 600 U.S. secondary schools. The authors also had census data, meaning that they could obtain characteristics of the communities in which the schools were located, along with self-report data from students, teachers, and principals on a range of different issues. The results were somewhat complex and varied between junior and senior schools and between teacher and student victimization. Generally, however, lower levels of student perception of fairness and clarity of rules related to higher victimization rates. The use of ambiguous sanctions (e.g., ignoring misconduct or lowering grades as a sanction), punitive attitudes on the part of teachers, and the presence of fewer resources also related to higher levels of teacher victimization.

In more recent work, Welsh (2001) investigated the effect of school climate and individual factors on a variety of outcomes: avoidance of certain areas in the school because of fear of victimization; offending (hitting, threatening, stealing, selling drugs in school); victimization at school; misconduct at school (being suspended or sent out of class as punishment); and safety at school. School climate was operationalized as the sum of five scales: respect for students, planning and action, fairness of rules, clarity of rules, and student influence. Above and beyond individual factors, respect for students, fairness of rules, clarity of rules, and student influence was each significantly related to lower levels of offending at school.

Multi-level studies employing hierarchical linear modelling (HLM)--allowing an investigation of effects at both the individual and the school level--have yielded similar findings to the earlier studies, which used individual-level data and OLS regression. For example, Bryk and Driscoll (1988) hypothesized that schools with a strong sense of community would have lower levels of misconduct and delinquency. They separated the effects of school composition (student background, minority composition of school, school social class, etc.) from aggregate individual variables. They found that school composition affected delinquency: larger classes and lower income levels were related to higher levels of delinquency. However, a strong sense of community within a school accounted for lower levels of problem behaviour.

Generally, then, both the earlier research by Rutter et al. (1979) and Gottfredson and Gottfredson (1985) and the later HLM research suggest that various aspects of the school climate work to inhibit delinquency. In reviewing the criminological literature on school effects on delinquency, Gottfredson (2001) concluded that

The simplest multilevel studies of school effects have generally examined effects on individual-level outcomes, controlling for individual demographic variables expected to predict the outcome ... These studies have demonstrated that something about where one goes to school matters for a wide range of student outcomes, including delinquent behaviour. (77) This article aims to further explore what that something is, using the concepts of instrumental and emotional support.

Instrumental and emotional support

"School climate" is a broad term and has therefore been conceptualized in numerous ways. As summarized above, for example, the delinquency literature has explored a wide range of different aspects of the school climate (e.g., clarity and fairness of rules, sense of community within the school). However, many would agree that the school climate is likely experienced at different levels (see Anderson 1982; Griffith 1999; Stockard and Mayberry 1992). For example, classrooms, though they are nested within schools, can have their own climates, which affect students in a variety of ways (Stockard and Mayberry 1992). In a review of the research on classroom climates, Stockard and Mayberry suggest that "effective classrooms appear to promote positive relationships among classroom members and have procedures oriented toward academic success" (1992: 33). These two aspects--relationships and academics--have also been applied to the broader school climate. Griffith (1999), for example, has suggested that the school climate facilitates the accomplishment of individual and group tasks (instrumental support) and also facilitates the development of social interactions, which, in turn, fulfil individual and group members' identity and interpersonal needs (expressive or emotional support).

Using perceptions from elementary school children, Griffith (1999) found that aspects of the school climate that were most immediate to the students (e.g., classroom factors) had the largest effects. More specifically, teachers' focus on pushing students to do well and helping with academic problems ("instrumental support" in the classroom), as well as student-teacher and...