Meta-Analytic Methods for Criminology (From Annals of the American Academy of Political and Social Science, Volume 578, November 2001, P 71-89, 2001, David P. Farrington, Brandon C, Welsh, eds. -- See NCJ-195740)

• NCJ Number: 195744
• Author(s): David B. Wilson
• Date Published: 2001
• Length: 19 pages
• Sale Source: Sage Publications, Inc
• Publication Format: Book (Hardbound)
• Publication Type: Report (Study/Research)
• Language: English
• Country: United States of America

Annotation

This article reviews the methodology of the quantitative data synthesis technique of meta-analysis (the statistical analysis of the results of prior research studies), finding that the inclusion of a meta-analysis in a systematic review has the capacity to produce the most rigorous summary assessment of the effectiveness of a criminological intervention based on a number of evaluation studies.

Abstract

This review of the use of meta-analytic methods to synthesize empirical relationship across studies, such as the effects of a specific crime prevention intervention on criminal offending behavior, reveals a robust empirical relationship but also identifies existing weaknesses in the knowledge base. Meta-analysis is found to avoid the shortcomings of the vote-count method by focusing on the size and direction of effects across studies rather than the statistical significance of individual effects. The framework of meta-analysis in terms of effect size, standardized mean difference, odds ratio, and correlation coefficient are discussed with recommendations for their use in meta-analysis. The analysis of meta-analysis data, including transformations and adjustment, statistical independence among effect sizes, inverse variance weight, mean effect size and related statistics, categorical analysis of effect sizes (the analog to ANOVA), meta-analytic regression analysis, fixed and random effects models, sensitivity analysis illustration (cognitive-behavioral programs for sex offenders), are discussed. The interpretation of meta-analytic findings in terms of attaching the term "significant" to an observed mean effect size; advantages and disadvantages of meta-analysis in terms of creating a solid foundation for future studies; and when not to do meta-analysis, such as to address broad theoretical issues, is covered here. In summary, it is emphasized that meta-analysis takes systematic reviewing a step further by quantifying the direction and magnitude of the findings of interest across studies. In part because it uses specialized statistical methods to analyze the relationship between finding and study features, meta-analysis can provide a firm foundation for future research. Also, meta-analysis results can be translated into summary statistics and be used to inform public policy regarding effective crime prevention efforts. Appendix, references