The near repeats of gun violence using acoustic triangulation data

Abstract

Near repeat signatures of gun violence have been generated exclusively from incidents reported to or discovered by police. However, many gun crimes go unreported. In Trenton NJ, it is believed as many as 60% of all such cases do not come to the police’s attention. This research explores near repeat qualities of gun-related violent crime in Trenton using an underutilized dataset, sounds of gunshots from ShotSpotter, allowing us to capture both reported and unreported gun-related violent crime. When examining reported and unreported gun violence, compared to solely relying on reported cases, near repeat signatures are smaller geographically and exhibit larger and more consistent statistically significant effects across temporal bands. Near repeat chains also occur more frequently and include more events. We caution against only using reported events in near repeat analyses. Improved measures of crime may change the character of near repeat events, research findings, and the operational response.