Abstract
Based on a theoretical model that emphasises the distinction between individual and contextual determinants of antisocial behaviour, the current study examined whether school attendance throughout adolescence acted as a protective factor for individuals at risk for criminal behaviour in early adulthood. Specifically, Lack of Control, an index of self-regulation which has previously been shown to predict later criminal behaviour, was expected to interact with early school leaving to predict self-reports and official records of criminal behaviour collected at age 21. Multivariate regression analyses revealed a significant three-way interaction between school attendance, self-regulation, and sex. Among males, after controlling for the effects of socioeconomic status and IQ, the main effects for Lack of Control and school attendance were found to be significant; additionally, the interaction between Lack of Control and school attendance was significant, indicating that the strength of the relation between Lack of Control and criminal outcomes was moderated by school attendance. The main effects for Lack of Control and school attendance were significant for females, but the interaction between Lack of Control and school attendance was not significant. The protective effect of school attendance among males could not be accounted for by differences in familial disruption or adolescent delinquency.