U.S. flag

An official website of the United States government, Department of Justice.

NCJRS Virtual Library

The Virtual Library houses over 235,000 criminal justice resources, including all known OJP works.
Click here to search the NCJRS Virtual Library

A Comprehensive Meta-Analysis of Randomized Evaluations of the Effect of Child Social Skills Training on Antisocial Development

NCJ Number
Journal of Developmental and Life-Course Criminology Volume: 7 Dated: 2021 Pages: 41-65
Andreas Beelmann ; Friedrich Losel
Date Published
March 2021
25 pages

This article presents a comprehensive meta-analysis of randomized controlled trials (RCTs) on the effects of child training on social skills and life course engagements.


Since developmental and life course criminology (DLCC) engages not only in correlational longitudinal research but also in programs of developmental prevention, within this context, child training on social skills plays an important role, so the present article is a comprehensive meta-analysis of randomized controlled trials (RCTs) on the effects of this type of intervention. The authors updated a meta-analysis on this topic Lösel & Beelman (Annals of the American Academy of Political and Social Science 587:84–109, 2003) to cover more recent studies while focusing specifically on aggression, delinquency, and related antisocial outcomes. From a systematic search of 1133 reports, we found 113 studies with 130 eligible RCT comparisons between a program and control group. Overall, 31,114 children and youths were included in these evaluations. Most interventions were based on a cognitive-behavioral approach. Overall, the mean effect was positive, but rather small (d = 0.25 using the random effect model). There were similar effects on aggression, delinquency, and other outcomes, but a tendency to somewhat stronger effects in behavior observations and official records than in rating scales. Most outcome measurements were assessed within 3 months or up to 1 year after training. Only a minority (k = 14) had follow-up assessments after more than 1 year. In the latter studies, mean effects were no longer significant. Indicated prevention for youngsters who already showed some antisocial behavior had better effects than universal approaches and (partially related to this) older youngsters benefited more than preschool children. There was much heterogeneity in the findings. Evaluations performed since our previous meta-analysis in 2003 did not reveal larger effects, but training format, intensity, and other moderators were relevant. Mean results are promising, but more long-term evaluations, replications, booster approaches, and combinations with other types of interventions are necessary to ensure a substantial impact on antisocial development in the life course. (Published abstract provided)