This meta-analysis studied the effectiveness of school violence prevention and treatment programs for reducing aggressive and disruptive behavior using Becker's (1988) techniques for analyzing standardized mean change effect sizes. In addition, the v...
This meta-analysis studied the effectiveness of school violence prevention and treatment programs for reducing aggressive and disruptive behavior using Becker's (1988) techniques for analyzing standardized mean change effect sizes. In addition, the validity of Becker's mean change method for this application was assessed. Results from a subset of studies analyzable under the mean change approach and the traditional group difference approach showed substantial agreement. The average mean change effect size for treatment groups differed from the average mean difference effect size only in the amount of change exhibited by comparison groups. Several threats to internal validity did not appear to be plausible alternative explanations for the results. In addition, a comparison of treatment groups from one-group pretest-posttest designs to treatment groups from control group designs revealed only slight differences. For this domain of research, the mean change approach appeared to be a viable alternative to the group difference approach to meta-analysis. In terms of program effectiveness, the most convincing evidence was found for two strategies: self-control or anger management programs and behavioral or classroom management interventions. Using prevalence estimates of school misbehavior, the effect size of .52 for behavioral treatments is equivalent to a reduction in misbehavior of 47% to 23%. In addition, treated subjects at higher risk levels showed greater change from pretest to posttest than lower risk subjects. And, children who participated in programs that were well implemented showed greater improvements than children who participated in programs that were less well implemented.