NCJ Number
209827
Date Published
May 2005
Length
341 pages
Annotation
This document is a detailed "how to" guide for analyzing race data obtained from vehicle stops made by police.
Abstract
Racial bias in policing is an issue that has plagued policing for many decades. Recent reform efforts have arguably reduced discriminatory practices in policing. It remains necessary, however, to analyze police behavior to ensure that discriminatory practices continue to decline. This guide provides a social science framework for measuring racial bias in police vehicle stops and outlines a variety of methods for use in the interpretation of vehicle stop data. "Benchmarking" is the method of analysis used to measure whether policing in a particular jurisdiction is racially biased. Benchmarking proves difficult because analysts must take into consideration a myriad of factors, such as the fact that racial/ethnic groups are not equally represented within the population. Following the introductory chapters, the guide presents step-by-step instructions for implementing the various benchmark methods recommended here. Data analysis guidelines for all benchmark methods are discussed and include descriptions of how to review data quality, the reasons for analyzing subsets of data, and how to match the numerator and the denominator. The details of benchmarking with adjusted census data, benchmarking with Department of Motor Vehicles data, benchmarking with data from blind enforcement mechanisms, observation benchmarking, and benchmarking with data for matched officers or matched groups of offenders are all described. Guidelines are offered for the post-stop analysis as well as for drawing conclusions from the results and using the results for police reform. Appendixes, references