NCJ Number
223600
Journal
Journal of Experimental Criminology Volume: 4 Issue: 2 Dated: June 2008 Pages: 117-135
Date Published
June 2008
Length
19 pages
Annotation
This study assessed the contribution that cost-benefit analysis could make to public policy decisionmaking.
Abstract
The study found that effect size was only weakly related to net benefits, as the analysis tested and rejected the hypothesis that cost benefit analysis does not contribute new information beyond the effect size alone. The rank order of net benefits and effect size were minimally correlated and effective analysis and cost benefit analysis led to different policy outcomes. Furthermore, the study found that two analytic methods would yield opposing policy recommendations for more than one in four interventions. These bivariate findings were supported by the results of multivariate models. Multivariate analysis was used in the study to try and isolate the effect of methodological variation. The paper questions whether undertaking a cost-benefit analysis provided additional information to policymakers as compared to an analysis solely of the effect of an intervention. However, a need for further research was cited to verify the accuracy of the standard errors on net benefit estimates, and recommended caution in the interpretation of these models. Data were derived from a literature review which identified 106 evaluations of criminal justice interventions that reported both an effect size and measures of net benefit. Data on net benefit and effect size were extracted from these studies. Tables, appendix, references