Several recent meta-analyses of universal school-based violence prevention studies indicate the overall positive impacts of these approaches on aggression. These studies, however, assess impacts on broadly defined measures of aggression. Furthermore, little research has analyzed the mechanisms through which these programs attempt to reduce overt aggressive behavior. The current study analyzed overall impacts on a more narrowly defined outcome--overt aggressive behavior--and identified associated mediators in 36 universal prevention studies conducted with kindergarten through fifth-grade students. Programs were associated with a significant, although small, reduction in overt aggression behavior. Three types of mediators were identified: measures of skill acquisition, social-cognitive processes, and classroom characteristics. Using MacKinnon's joint significance test to test for mediation, four measures of skill acquisition, two measures of social cognitive, and one measure of classroom characteristics were identified as significant mediating variables. Implications for the design of effective violence prevention programs and mediators to assess in future research are discussed. (Contains 3 tables, 1 figure, and 1 footnote.)