Family, school and friends: Correlates of Canadian youth outcomes

Aperçu

Résumé (français)

Veuillez noter que les résumés n'apparaissent que dans la langue de la publication et peuvent ne pas avoir de traduction.

Résumé (anglais)

This paper empirically examines the joint correlates of academic performance and criminal behaviour of Canadian 12- to 15-year old boys and girls. Previous researchers have studied various child or youth outcomes, for example, physical health, academic performance, behaviourial outcomes, and labour market outcomes in young adulthood. However, the literature on correlates of juvenile criminal behaviour using Canadian data is very thin. This paper aims to fill in this gap. Using the National Longitudinal Survey of Children and Youth (NLSCY), I consider four categories of potential correlates: personal and family characteristics, neighbourhood characteristics, school environment, and public policies. Some main findings are: (1) School quality measured by “in-class cooperation through group activities” is linked to better academic performance and less crime for boys and girls, while conventional school quality measures, such as class size or teacher’s education level, are not significant; (2) Observable neighbourhood characteristics, such as median income, racial heterogeneity and age composition, do not seem to matter; (3) There is a strong “peer” group effect for both academic performance and criminal activities; and (4) Consistent with existing literature, family background is very important for Canadian youth outcomes. The contribution of this paper is three-fold. first, to the author’s knowledge, this is the first study that considers the relationship between school environment and Canadian youth outcomes using the NLSCY. Second, this study further adds to existing literature by including an extensive list of other explanatory variables in addition to school environment, thus mitigating potential omitted variable bias. finally, considering the joint correlates of academic performance and participation in crime using multivariate probit regression analysis also provides a new perspective.