[Paper Review] Policing Chronic and Temporary Hot Spots of Violent Crime: A Controlled Field Experiment

Dylan Fitzpatrick, Wilpen L. Gorr+1 more|arXiv (Cornell University)|Nov 11, 2020

Crime Patterns and Interventions17 references4 citations

TL;DR

This controlled field experiment evaluates proactive policing in Pittsburgh by targeting both chronic and temporary hot spots of serious violent crime using predictive analytics. It finds a 25.3% reduction in Part 1 Violent crimes and a 39.7% reduction in crimes against non-white victims, with foot patrols showing significant crime reduction and no evidence of crime displacement or over-policing of minorities.

ABSTRACT

Hot-spot-based policing programs aim to deter crime through increased proactive patrols at high-crime locations. While most hot spot programs target easily identified chronic hot spots, we introduce models for predicting temporary hot spots to address effectiveness and equity objectives for crime prevention, and present findings from a crossover experiment evaluating application of hot spot predictions to prevent serious violent crime in Pittsburgh, PA. Over a 12-month experimental period, the Pittsburgh Bureau of Police assigned uniformed patrol officers to weekly predicted chronic and temporary hot spots of serious violent crimes comprising 0.5 percent of the city's area. We find statistically and practically significant reductions in serious violent crime counts within treatment hot spots as compared to control hot spots, with an overall reduction of 25.3 percent in the FBI-classified Part 1 Violent (P1V) crimes of homicide, rape, robbery, and aggravated assault, and a 39.7 percent reduction of African-American and other non-white victims of P1V crimes. We find that temporary hot spots increase spatial dispersion of patrols and have a greater percentage reduction in P1V crimes than chronic hot spots but fewer total number of crimes prevented. Only foot patrols, not car patrols, had statistically significant crime reductions in hot spots. We find no evidence of crime displacement; instead, we find weakly statistically significant spillover of crime prevention benefits to adjacent areas. In addition, we find no evidence that the community-oriented hot spot patrols produced over-policing arrests of minority or other populations.

Motivation & Objective

• To evaluate the effectiveness of proactive policing in reducing serious violent crime by targeting predicted hot spots, including both chronic and temporary locations.
• To assess whether predictive policing strategies can improve crime prevention while maintaining racial equity and avoiding over-policing of minority communities.
• To examine whether crime reduction in hot spots leads to spillover benefits in adjacent areas or results in crime displacement.
• To compare the impact of foot patrols versus car patrols in reducing violent crime within targeted hot spots.
• To evaluate whether community-oriented proactive policing reduces crime without increasing arrests of racial or ethnic minorities.

Proposed method

• Conducted a crossover experimental design in Pittsburgh, PA, assigning uniformed patrol officers to weekly predicted hot spots across 12 months.
• Used predictive modeling to identify both chronic hot spots (persistent high-crime locations) and temporary hot spots (short-term crime surges) based on historical crime data.
• Applied a difference-in-differences approach to compare crime counts in treatment hot spots versus control hot spots, with adjustments for week-level and zone-partition fixed effects.
• Distinguished between foot patrols and car patrols to assess differential impacts on crime reduction.
• Measured outcomes using Part 1 Violent (P1V) crime counts (homicide, rape, robbery, aggravated assault) and analyzed spillover effects in adjacent zones.
• Conducted OLS regression models with robust standard errors to estimate treatment effects, including interactions between treatment and spatial proximity to hot spots.

Experimental results

Research questions

• RQ1 Does proactive policing targeting predicted hot spots reduce serious violent crime in Pittsburgh?
• RQ2 Are temporary hot spots more effective for crime reduction than chronic hot spots, and how do their impacts compare?
• RQ3 Does foot patrol produce greater crime reduction than car patrol in targeted hot spots?
• RQ4 Is there evidence of crime displacement to areas outside of treated hot spots?
• RQ5 Does proactive policing lead to over-policing of minority populations, particularly in terms of arrests?

Key findings

• A 25.3% reduction in Part 1 Violent (P1V) crimes was observed in treatment hot spots compared to control hot spots over the 12-month period.
• A 39.7% reduction in P1V crimes involving African-American and other non-white victims was found in treatment areas, indicating strong racial equity in outcomes.
• Foot patrols produced statistically significant reductions in violent crime, while car patrols did not show a significant effect.
• No evidence of crime displacement was found; instead, weakly significant spillover effects were observed in areas adjacent to hot spots.
• There was no evidence that community-oriented hot spot patrols led to over-policing of minority populations or increased arrests of non-white individuals.
• Temporary hot spots showed a higher percentage reduction in P1V crimes than chronic hot spots, though they prevented fewer total crimes due to their short-lived nature.