|This paper assesses the utility of meta-analysis by comparing its yield to that of qualitative literature review in ascertaining extant knowledge on the relationship between inequality and crime. The data for this analysis are drawn from thirty-four inequality and crime aggregate data studies reported by Hsieh and Pugh (1993), published between 1974 and 1991 and covering the 1970s and 1980s. These studies obtained multiple estimates of inequality or unemployment crime relationships, which could then be summarized as average correlations obtained under different methodological conditions. In contrast, the qualitative approach to research synthesis is examplified by Chiricos's (1987) examination of the direction and significance of the unemployment effects on crimes according to methodological conditions.
Applying the qualitative approach to the inequality - crime relationship shows complex and contradictory results. Both bivariate and multivariate analyses yield similar proportions of positive-significant and negative-significant inequality - general violent crime relationships. However, results are more varied when specific violent offenses were examined. With respect to property crime, multivariate estimates are yet more ambiguous.
This study concludes that meta-analysis is useful in obtaining a general summary estimate of the strength of the relationship between a dependent and an independent variable. However, until further research helps us to know how these important and theoretically conflicting factors influence crime, meta-analysis cannot substitute for the rich knowledge obtained by qualitative analysis.
(Return to Program Resources)