Body-worn cameras, criminal convictions, and the choice of denominator: a research note

Abstract Objectives To examine the sensitivity of research on the downstream effects of body-worn cameras (BWC) to the denominator used to calculate effect sizes. Methods I re-examine results from a recent systematic review and meta-analysis on the downstream effects of BWCs and calculate effect sizes based on two denominators: (1) All cases, arrests, or incidents included in the study and (2) prosecuted cases only. I then estimate meta-analytic models under both specifications and compare the results. Results The downstream effects of BWCs are considerably larger and more heterogeneous when estimated relative to all cases or arrests. Models based on prosecuted cases only suggest small but significant decreases in guilty pleas and combined guilty outcomes, with little between-study variability. Conclusions Through the use of broad denominators, extant research often compounds the effects of BWCs on criminal convictions and guilty dispositions with prior downstream outcomes. BWCs may produce increases in guilty dispositions, but this effect appears to occur through more frequent prosecutions. Conditional on prosecution, existing studies provide consistent evidence that BWCs have little impact on guilty dispositions, and may even result in a small exoneration effect.