Case studies of predictive analysis applications in law enforcement

Law enforcement executives and policy makers continuously seek out effective strategies to reduce crime. Reducing crime reduces social harm, improves community resilience, and therefore improves homeland security. Before investing in a crime control strategy, police leaders must know if the effectiveness of that strategy has been validated. Predictive policing is one such strategy in use that relies on mathematical algorithms to forecast probable future crime locations and the application of interventions to interdict or prevent crime in those locations. In this thesis, theories and methodologies behind predictive policing are described, and the case study method is used to review current predictive policing practices. The research finds that despite the conventional wisdom that a correlation exists between the implementation of a predictive policing program and a reduction in crime, no evidence indicates that a direct cause and effect relationship exists. This thesis provides law enforcement executives and policy makers with objective research on the effectiveness of predictive analysis in reducing crime and provides recommendations for those evaluating whether to invest time and resources into a predictive policing program.