Measuring and interpreting repeat victimization using police data: A study of burgaries in Charlotte, North Carolina

This study of repeat victimization and crime prevention tackles issues relating to the measurement and interpretation of police data. In a study of repeat burglaries in Charlotte, NC, published in this series, LeBeau and Vincent (1997) framed their discussion as a critique of our work on preventing repeat victimization. We reanalyze and reinterpret the data and findings from that study. First, we explain why the police data significantly understated the true rate of repeat burglaries in Charlotte. Second, we reanalyze the published data to show that, nevertheless, there were at least nine times more repeat burglaries than would occur by chance and that, after a first burglary, the predictability of repeats rises dramatically with each subsequent burglary. Third, we explain why the policy conclusions of LeBeau and Vincent were misleading in relation to preventing repeat burglary. Fourth, we explain why their policy conclusions were misleading with respect to the potential use of burglar alarms as part of a prevention strategy. Fifth, we list some of the relevant literature that the study should have considered more closely. We conclude that their 1997 study did a disservice to the study of repeat victimization and crime prevention, and we reassert that the prevention of repeat burglaries could be a useful component of crime prevention strategy in Charlotte. More generally, the need for academic studies to consider the methodological issues relating to the measurement of repeat victimization is not sufficient reason to avoid or delay practical crime prevention efforts. We conclude with a "hit rate challenge" that crime prevention research should seek to identify predictors that are more efficient than, and at least as practical as, prior victimization.