Measuring recurrent victimization: evaluating operationalization strategies and predictors using the Crime Survey for England and Wales

There is little consensus on how recurrent victimization should be operationalized. This study evaluates alternative measurement strategies and identifies predictors of recurrent victimization using a meta-analytic synthesis of analytic approaches applied to the Crime Survey for England and Wales. Results show that defining recurrent victimization using a Top 10% binary threshold and estimating logistic regression models can lead to biased conclusions. In contrast, operationalizations based on experiencing two or more victimization types or incidents perform better when analyzed using bivariate probit models. Count-based measures, particularly total victimization counts across crime types, also perform well when estimated using negative binomial or zero-inflated negative binomial models. Overall, the findings suggest that categorization-based approaches should be theoretically informed and paired with bivariate probit models, while analyses of victimization volume are better suited to count-based frameworks. Across all approaches, mental health conditions consistently emerge as the strongest correlate of recurrent victimization.