Police Stops, Crime Prevention, and Community Reaction: A Randomized Field Experiment at Violent Crime Hot Spots

The applicant proposes to conduct a randomized experiment to test whether an enhanced version of pedestrian stops, focused on violent crime hot spots and informed by constitutional and procedural justice principles, can significantly reduce crime in these spots while enhancing perceptions of procedural justice among those stopped, and increasing community opinions about police. The applicant proposes to identify the top 1% of the most violent street segments (120 total segments) in Phoenix (AZ). The researchers will then block, randomize, and select 60 segments (including the intersections) as treatment and 60 as control groups. They will also conduct a preliminary analysis to ensure no other special interventions are occurring there. For the treatment group, a select unit of five (5) officers will be chosen. These officers will receive a special, 5-day, 40-hour, training focused on how and when to conduct stops so that they are focused on crime reduction while remaining constitutionally and procedurally just. That is, they will train to distinguish between beginning a stop (based on reasonable suspicion) as opposed to a frisk (based on a reasonable suspicion and a belief that the suspect is armed/dangerous). After their training, they will go into the field for a 10-hour shift with a supervisor to conduct stops as a test of their training. I If additional training is needed at that point, or is identified to be needed at any point during the study, they will receive additional training. Upon finishing this field test, the officers will be sent into the treatment areas where they will be encouraged to use pedestrian stops as their primary tactic in conducting directed patrol. The control group will be business as usual with no dedicated officers.
There are four (4) main sources of data generated by this application. After each stop, officers will be required to fill out a report detailing administrative measures of crime; surveys of those stopped; and surveys within the communities where the stops occur. Three main measures will be used to test this intervention:. First, the applicant proposes measuring crimes in the treatment group versus the control group by comparing the groups both before and during the intervention. Second, the applicant will compare satisfaction with the encounters and perceived police authority, both measured using contact surveys with persons stopped in the control and treatment segments. Third, pre-/post-intervention community surveys will be conducted, focusing on legitimacy and efficacy of police work. Finally, the applicants propose an additional measure of fidelity by reviewing 240 (120 treatment and 120 control) body-worn camera videos of conducted stops.

Note: This project contains a research and/or development component, as defined in applicable law," and complies with Part 200 Uniform Requirements - 2 CFR 200.210(a)(14). ca/ncf