NCJ Number
169028
Date Published
1997
Length
309 pages
Annotation
The impact on recidivism of interventions focused on the individual was examined by means of a meta-analysis of 150 studies that used a randomized experimental design, were written in English and published during 1950-93, and included a quantifiable outcome measure of crime.
Abstract
The studies examined interventions that included treatment programs (counseling), deterrence strategies (arrests), and juvenile delinquency prevention programs (casework with at-risk youth). Each study was coded using a 196-item instrument. Intercoder reliability was assessed on a random subset of studies and produced an 80-percent rate of agreement. This analysis revealed that 115 interventions were rehabilitative in focus, 23 were oriented to deterrence, and 9 were juvenile delinquency prevention programs. Cohen's effect size served as the common measure. Findings indicated that the global mean effect size was highly unstable. The effect size for rehabilitation programs was much larger than that for deterrence and prevention. Further analysis using sample-size weighting reduced the global mean for rehabilitation programs to a level smaller than the deterrence effect. Further analysis revealed that rehabilitation studies with small samples of 10-100 participants reported much larger effects than experiments with 300 or more participants. In addition, experiments with juveniles were more effective than those for persons age 18 or over. Recommendations for strengthening meta-analysis and for future research, figures, tables, footnotes, appended background information and author biography, and approximately 400 references