When Is a “War” a “Wave?” Two Approaches for the Detection of Waves in Gang Homicides

Abstract

Gang violence and gang “wars” are often described as coming in waves, but little empirical work has been conducted to distinguish between actual “waves” of violence and the more common ups and downs that trends in homicides typically go through. We propose two approaches for the detection of waves of gang-related homicides in the Lower Mainland of British Columbia for a time period (2006–12) when these were considered to occur at a high rate: (1) the monthly waves approach, whereby monthly crime data are used to map the trends in gang-related violence, and (2) the micro-approach, whereby crime waves are detected by examining significant deviations from the mean number of days between homicides. The results show that four distinct monthly waves could be detected between 2006 and 2012, each capturing the peak moments of known gang conflicts. The micro-approach led to the discovery of 12 waves, allowing for a more sophisticated understanding of the trends in gang violence. While the identification of “trigger events” before the rise of a wave was relatively straightforward for the four monthly waves, not all 12 micro-waves could be associated with a clear trigger event using open source data. The two approaches should be used in complementarity for a meaningful and accurate understanding of trends in gang violence.