Assessing the Relative Strength of Macro-Level Predictors of Crime: A Meta-Analysis

Travis C. Pratt, Rutgers University
Francis T. Cullen, University of Cincinnati

Macro-level (or "ecological") theory and research emerged (or "re-emerged") and has since earned sustained criminological attention. Prompted by new empitical and theoretical developments, over 200 empirical studies have been conducted and published in academic journals in an effort to uncover the correlates of aggregate levels of crime. The present study subjected the body of macro-level criminological literature to a "meta-analysis" -- or "quantitative synthesis" -- to determine the relative effects of thirty-one macro-level predictors of crime assessed across empirical studies. The results indicate that macro-level indicators of "concentrated disadvantage" are among the strongest and most stable predictors of crime across empirical studies. These include racial heterogeneity (measured as the percent non-white and/or the percent black), poverty, and family disruption. Conversely, variables related to criminal justice system dynamics (e.g., policing effects, the effects of "get tough" policies) are among the most consistently weak predictors of crime. The implications of the analysis for future research and public policy development are discussed.

(Return to Program Resources)

Updated 05/20/2006