Crime Mapping and Policing

Abstract

The empirical truism that crime clusters geographically has increasingly informed police practice in the first two decades of the 21st century. This has been greatly aided by crime mapping, which can be thought of as a summary term for the geographic analysis of data related to crime and disorder. Crime analysts use a range of mapping techniques to describe and explain patterns in data, being mindful of the unique qualities of geographically referenced data. These include point pattern maps, thematic maps, kernel density estimation, and GI* statistic maps. Rate maps are also used when the aim is to understand the risk of victimization. Crime maps can be used to identify hot spots for targeted action, identify recently committed crimes that might form part of a linked series, monitor the impact of police initiatives, and aid understanding of crime problems. Environmental criminology theory is often used for the latter of these purposes, as this can explain spatiotemporal patterns revealed through crime mapping and assists in understanding the mechanisms driving the patterns. The logic for organizing police efforts around geographical crime concentration is instinctive: by focusing on a small number of locations with high crime volumes, police can be more focused and hence effective. Thus, “hot spots policing” has become business as usual in many jurisdictions, focusing attention at small units of geography. Hot spot policing has, in many countries in the early 21st century, evolved into “predictive policing,” which is a data-driven crime forecasting approach. However, this approach has attracted strong criticism for further entrenching bias in the criminal justice system. Critics argue that the data sets on which the predictive policing algorithms are founded are racially biased and perpetuate police attention on communities of color. This serves to highlight the systematic biases that can be present in crime-related data, and professional crime analysts use a range of approaches to mitigate such biases, such as triangulating data sources.