|
Activity Number:
|
49
- Statistical Measurements of Social Issues and Trends
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, August 3, 2020 : 10:00 AM to 2:00 PM
|
|
Sponsor:
|
Social Statistics Section
|
|
Abstract #313743
|
|
|
Title:
|
Racial Disparities in Policing: A Variance Decomposition Approach
|
|
Author(s):
|
Mikaela Meyer* and Amelia M Haviland
|
|
Companies:
|
Carnegie Mellon University and Carnegie Mellon University
|
|
Keywords:
|
hierarchical modeling;
equity;
law enforcement;
racial disparities
|
|
Abstract:
|
Previous studies have analyzed racial disparities in police use of force in a city using various statistical methodologies. Understanding whether these disparities vary across units of geography would help determine which disparity-reducing interventions should be prioritized within a community. The goal of our research is to understand at which geographic level—precinct or census tract—the variance in potential racial disparities in police use of force is largest. We use hierarchical logistic regression models to analyze the probability of more severe types of force being used against individuals of a given race. Through our collaboration with the Center for Policing Equity, we use data from 11,224 use-of-force incidents occurring between 2008 and 2019 in a midwestern city that represent all five of the city’s precincts and 104 of the city’s census tracts. After controlling for various census tract characteristics using American Community Survey data, we find the variance in the level of these probabilities is higher across census tracts. Yet the variance in the disparities of these probabilities between Blacks and Whites is higher across precincts.
|
Authors who are presenting talks have a * after their name.
