LONG‐TERM TRENDS IN CRIMES OF VIOLENCE (COMMENT ON COONEY, 2003)*

Empirical and theoretical issues posed by Mark Cooney's (2003) analysis of long‐term trends in crimes of violence are discussed.     

THE INFLUENCE OF EARLY SCHOOL PUNISHMENT AND THERAPY/MEDICATION ON SOCIAL CONTROL EXPERIENCES DURING YOUNG ADULTHOOD

The use of suspensions and expulsions by American public school administrators has increased dramatically over the past 40 years. Meanwhile, a growing number of childhood misbehaviors have been diagnosed by doctors as medical conditions and are being treated with therapy or medication. As these trends develop at different rates for boys of different racial and ethnic groups, the connection between childhood and adult social control remains untested empirically. By using a prospective panel of 3,274 White, Black, and Hispanic males (15,675 person‐years) and multilevel logistic models, I examine whether and how school punishment and/or the use of therapy or medication during childhood contributes to involvement in the criminal justice or mental health systems during young adulthood. The findings suggest that school punishment is associated with greater odds of involvement in the criminal justice system but not the mental health system. The use of therapy and/or medication during childhood is associated with higher odds of involvement in the mental health system but not the criminal justice system. Finally, although the relationship between school punishment and involvement with the criminal justice system is similar for White, Black, and Hispanic men, the relationship between medicalized social control during childhood and young adulthood is stronger for Whites than for non‐Whites.     

UNSTRUCTURED SOCIALIZING,COLLECTIVE EFFICACY,AND VIOLENT BEHAVIOR AMONG URBAN YOUTH*

Relying on extensions of routine activities and social disorganization theories, we examine whether 1) neighborhood social characteristics shape opportunities for the development of unstructured socializing with peers among adolescents, 2) whether unstructured socializing leads to an increase in violent behavior within urban communities, and 3) whether neighborhood collective efficacy modifies the impact of unstructured socializing on violence. The study outlined in this article uses three waves of data from the Project on Human Development in Chicago Neighborhoods Community Survey and Longitudinal Cohort Study. Results from multilevel linear models suggest that neighborhood collective efficacy supports the development of unstructured socializing with peers. Multilevel Rasch models of violent behavior indicate that, consistent with previous research, unstructured socializing is a powerful predictor of violence. Collective efficacy exerts an independent influence on violent behavior and attenuates the effect of unstructured socializing on this outcome.