Investigative Stops and Neighborhood Change: A Spillover Analysis of Chicago Police Beats

Abstract

This research tests for a spillover effect in the association between discretionary police contact and neighborhood economic and racial change at the police beat level in Chicago, Illinois. Spatial Durbin lag and error models are fit for 269 police beats, using investigative stop report data from 2016 to 2019 and American Community Survey estimates for 2006-2010 to 2015-2019. Two outcomes are modeled, the logged investigative stop count and the contraband recovery rate among searches, on an economic change index, a racial turnover index, change in racial concentration, logged violent crime, and logged population, each entering directly and as a spatial lag. Economic change in a beat’s surroundings is negatively associated with its stop volume, a total impact of -0.314 with a standard error of 0.088, but that association does not survive adjustment for Chicago’s 22 police districts. Beat adjacency and district membership are not separable, since two thirds of a beat’s neighbors share its district. No prior spillover study has made that adjustment, which is this project’s principal contribution. For planning, this work examines how discretionary police contact and neighborhood change interact along police beat and district lines, boundaries usually understood as outside the profession’s remit.