Abstract

The aim of the current project is to examine the extent to which two aspects of adolescents’ social functioning with peers (i.e., aggression and likeability) will predict two forms of adult criminal behavior (property and violent crimes). A second purpose is to determine whether these predictive associations vary as a function of neighborhood-level factors (socio-ecological disadvantage).
The sample was drawn from the Concordia Longitudinal Risk Project, first initiated in 1976. The current sample included 2,497 fourth- and seventh-graders drawn from mixed-sex classrooms in schools located in working class neighborhoods in Montreal, Quebec, Canada. Those students were first screened on measures of aggression and likeability via the Pupil Evaluation Inventory (PEI; Pekarik et al., 1976). Adult criminality information was obtained from the open access database of arrest and conviction records in Montreal. These measures have been coded according to whether the acts were perpetrated against property (e.g., breaking and entering, theft) or against people (e.g., murder, assault, kidnapping). A weighted neighborhood score was created based on four school neighborhood conditions (e.g., proportion of unemployed people, number of single parents, number of people with less than grade 9 education, and number of people in the neighborhood earning less than 10K/year).
Main effects of aggression were observed on both property (β = .12, SE = .05, t = 2.50, p = .01) and violence against people (β = .07, SE = .02, t = 2.47, p = .01) outcomes, with stronger effect for boys than for girls. Also, a negative association was observed between likeability and adult criminal convictions. This finding was moderated by grade, with stronger negative effect for 4th graders than 7th graders. Moreover, peer status was found to moderate the association between aggression and adult criminality. That is, aggressive youth with higher peer status were found to commit less adult criminal acts than aggressive youth with low status. There were no observed neighborhood effects on the association between childhood aggression and adult criminality.