Effects of Child Skills Training in Preventing Antisocial Behavior

1
Effects of Child Skills Training in Preventing
Antisocial Behavior

• By Friedrich Losel Andreas Beelmann
• Angelique Marshall
• Radford University

2
Purpose

• To analyze the preventive effects of child skills
  training on antisocial behavior and related
  outcomes.
• provide a basis for a differentiated evaluation
  of the preventive potential of child skills
  training
• permit conclusions on the practical
  implementation of such programs
• offer perspectives for practice, policy making,
  and research

3
Literature Review

• All retrievable published or unpublished reports
  in the English or German Languages that had
  appeared no later than 2000
• Methods
• Electronic Sources- keywordsPrevention
  antisocial behavior social skills training
  evaluation meta-analysis childhood and
  adolescence
• Psyc-Info
• Medline
• Eric
• Dissertation Abstracts

4
Literature Review

• References given in existing reviews were checked
  systematically
• References given already identified primary
  studies were analyzed for further relevant
  publications
• Studies
• Approximately 851 found
• 84 met eligibility criteria that included 135
  treatment/control group comparisons

5
Studies to Include

• Must contain an evaluation specifically
  addressing a social training program for the
  prevention of antisocial behavior in children and
  youth
• Randomized control group design
• Pre-intervention and post-intervention data had
  to be available
• Age zero to eighteen years
• Focus on prevention
• Outcome measures data had to be reported in
  sufficient detail to permit an adequate
  computation

6
Potential ModeratorsIndependent Variables

• Publication year
• Sample size
• Type of randomization
• Type of treatment
• Trainers
• Age
• Type of prevention

7
Potential ModeratorsDependent Variables

• Experimental design treatment and control group
• Specifically addressing a social training program
  for prevention of anti-social behavior
• Post-intervention measurements
• Follow-up measurements

8
Results for Dependent Moderators
Outcome Measurements N Fixed Model d
Post-intervention total 126 .26 Significant from 0 Heterogeneity
Antisocial behavior 80 .17 Significant from 0 Heterogeneity
Social skills 61 .29 Significant from 0 Heterogeneity
Social-cognitive skills 57 .29 Significant from 0 Heterogeneity
9
Results for Dependent Moderators
Outcome Measurements N Fixed Model d
Follow up total 34 .14 Significant from 0 Heterogeneity
Antisocial behavior 20 .06 Heterogeneity
Social skills 16 .13 Significant from 0 Heterogeneity
Social-cognitive skills 12 .33 Significant
10
Results for Independent Variable Moderators

• Low sample size in many of the studies
• Performed multiple significance testing without
  alpha adjustments
• Some effects are based on only a few studies
• Random model is not very sensitive for moderator
  effects
• Moderators are to some extent confounded

11
Conclusion

• Studies demonstrate a positive overall effect
  that is small but robust
• Cognitive-behavioral programs targeting high-risk
  youngsters who already exhibit some behavioral
  problems seem to be particularly effective.