Search

Co-occurring social-emotional problems are associated with increased risk of aggression and substance use. However, few studies examine their configurational patterns. This study identifies patterns of co-occurring social skills, anxiety, learning, and conduct problems among 2,632 urban boys at entry into sixth grade, and their related aggression and substance use trajectories through eighth grade. Latent class analysis revealed four patterns at school entry: "low-all," "poor social skills," "positive social skills," and "high all." Findings point to important variation in risk. Problem behaviors increased the least through middle school for the "low-all" pattern. The "positive social skills" pattern had an average increase, while the "poor social skills" pattern had higher levels of problem behaviors in sixth and seventh grade. The "high all" showed the fastest increase in problem behaviors and the highest levels in eighth grade. Discussion focuses on implications for a multi-tiered school-based system of supports for behavioral risk management.

This study examined the direct effects of beliefs about aggression and nonviolence on physical aggression and their role as protective factors that buffer adolescents from key risk factors in the peer, school, and parenting domains. Multilevel analyses were conducted on data from 5,581 adolescents representing two cohorts from 37 schools in four communities collected at the beginning and end of the sixth grade and at the end of the following 2 school years. Individual norms for aggression at Wave 1 moderated relations of delinquent peer associations and parental support for fighting with physical aggression. Self-efficacy for nonviolence at Wave 1 moderated relations of school risk, delinquent peer associations and parental support for fighting with physical aggression. There was clearer evidence for protective effects for self-efficacy for nonviolence for girls than for boys. (Contains 4 tables and 2 figures.)

To prevent youth violence, the GREAT Families program was implemented with a selective sample of 1,196 families of sixth-grade children from low-income schools in 4 cities making the transition to adolescence. To assess intervention effects, we used pre- and posttest data to estimate a structural model to test the hypothesis that random assignment to the intervention would predict changes in parenting practices, which in turn would predict changes in exposure to violence (i.e., a mediational model). We found that participation in GREAT was significantly related to changes in parenting practices and these changes significantly predicted changes in violence exposure. Furthermore, adolescents who reported greater increases in exposure to violence also tended to experience greater decreases in parental monitoring, discipline, and involvement. Implications for future research and prevention efforts are discussed. [ABSTRACT FROM AUTHOR]