Meta-Analysis and Systematic Review of the Effectiveness of School-Based Programs to Reduce Multiple Violent and Antisocial Behavioral Outcomes

Abstract

Early models of violence prevention targeted a single outcome by intervening in a single suspected causal pathway. This limited approach has increasingly fallen out of favor, as evidence has accumulated that violent and antisocial outcomes are driven by a highly inter-correlated cloud of potential risk factors. Policy makers and consumers are often interested in whether or not a program actually prevents or reduces the problem behaviors that it is designed to address. The current study combined the methods of a traditional meta-analysis and traditional systematic review to examine the evidence of eff ectiveness of school-based programs in simultaneously reducing both violent and antisocial behavioral outcomes. Overall, none of the programs report evidence of being eff ective for all outcomes, and only one successfully impacted more than one distal outcome. When considering all forms of evidence, no program showed uniformly positive evidence across all outcomes and domains considered. Implications for science and practice are discussed. Increasingly, prevention science is moving towards integrated models of prevention. Early models of violence prevention targeted a single outcome by intervening in a single suspected causal pathway (Domitrovich, Bradshaw, Greenberg, Embry, & Ialongo, 2009). This limited approach has increasingly fallen out of favor, as evidence has accumulated that violent and antisocial outcomes are driven by a highly inter-correlated risk factors (Wei, Loeber, & White, 2004). Prevailing evidence suggests that there is a complex interplay of predisposition, training and skills, and environmental stimuli that predispose youth to acting or responding inappropriately (Silberg, Rutter, D’Onofrio, & Eaves, 2003). Theoretically, each individual receiving an intervention has a unique cluster, or at least uniquely weighted cluster, of risk factors. That is, within individual,