Criminal Justice Faculty Publications and Presentations

A meta-analysis of the correlates of turnover intent in criminal justice organizations: Does agency type matter?

Document Type


Publication Date




  • Provides a comprehensive, meta-analytic, review of the turnover intent literature.

  • Covers multiple domains of the justice system including police and corrections.

  • Exposes diversity of variables and measurements used across/within domains.

  • Reaffirms substantive impact of job satisfaction on turnover intentions.

  • Encourages future studies to be more consistent in use of organizational variables.



The study synthesizes the literature on turnover intentions to assess what domains (e.g., personal characteristics, work environment, and job attitudes) account for the strongest association with turnover intent, what are the characteristics of these relationships, and how do these relationships differ by criminal justice practitioner type.


The current study utilizes a systematic review to obtain studies for conducting a meta-analysis. The researchers utilized the r family/correlation coefficient. Studies were weighted by sample size, correlations converted to Fisher’s z, analyses performed, and results converted back to r for interpretation.


In terms of the individual predictors for law enforcement, the five strongest variables included alternative job search behavior, job satisfaction, psychological distress, emotional exhaustion, procedural and distributive justice. The five strongest predictors of turnover intent for institutional corrections were normative commitment, emotional exhaustion, depersonalization, affective commitment, and job satisfaction. The five strongest predictors for community corrections included experience, alternative job search behavior, perceptions of coworkers, growth need strength, and job satisfaction.


The results of the meta-analysis confirmed the domain of personal characteristics, overall, has the weakest association with turnover intent whereas work environment and job attitude domains consistently display moderate-to-large effects for both law enforcement and corrections.


Original published version available at

Publication Title

Journal of Criminal Justice