Objective: The aim of this article was to provide data about the effects of child sexual abuse prevention programs. A more specific aim was to estimate the contribution of potential moderator variables such as age, program duration, or sample size to effect size. Method: A meta-analytic approach was used to calculate post-test and follow-up effect sizes of 16 evaluation studies of school programs aimed at the prevention of child sexual abuse victimization. Tests of categorical models were used in the analysis of moderator variables. Multiple regression analysis was used to determine their association with effect sizes. Results: Significant and considerable mean post-intervention (d = .71) and follow-up (d = .62) effect sizes were found, indicating that victimization prevention programs are successful in teaching children sexual abuse concepts and self-protection skills. Intervention characteristics such as duration and content of the program, and child characteristics such as age and SES were important moderators of effect size. Conclusions: Our findings corroborate and refine the positive conclusions of traditional narrative reviews. Programs that focus on skill training, allowing sufficient time for children to integrate self-protection skills into their cognitive repertoire, are to be preferred. Future evaluation research should focus on transfer of training.