NCJ Number
212815
Journal
Sexual Abuse: A Journal of Research and Treatment Volume: 17 Issue: 4 Dated: October 2005 Pages: 357-373
Date Published
October 2005
Length
14 pages
Annotation
After describing general approaches to integrating the findings of single studies into a summary of all the findings "meta-analysis," these methods are applied to two social programs whose effectiveness has been uncertain based on separate evaluations: treatment for sex offenders in the criminal justice system and rape prevention programs for college women.
Abstract
Using meta-analysis procedures, this study found that the cumulative weight of individual studies has shown that sex offenders treated in the criminal justice system were at lower risk of sexually reoffending than untreated sex offenders. A meta-analysis of evaluations of rape prevention programs for college women found that they had a small beneficial effect and that female-only programs were more effective than programs that included both men and women. The authors note that these meta-analyses of single evaluations of these two types of programs yielded conclusions about their effectiveness that none of the single studies could provide. This confirms the importance of meta-analyses of multiple single studies of similar subjects. The following general steps in meta-analysis are explained in this paper: the identification of previous relevant studies; computation of the confidence interval for the effect size in previous studies; computation of the confidence interval for the effect size in the current study; comparison of the current results with previous results; and computation of the confidence interval for the new effect size that includes past and current findings. 2 figures, 3 tables, and 53 references