Searching for the Denominator: Problems with Police Traffic Stop Data and an Early Warning System Solution

Abstract

In response to widespread allegations of racial and ethnic discrimination in traffic stops, a practice that has been labeled “racial profiling,” law enforcement agencies are now collecting data on traffic stops that include the race or ethnicity of drivers. Interpreting these data to determine whether a pattern of race discrimination exists poses enormous difficulties. Specifically, it is not clear what baseline (often referred to as “the denominator”) should be used to assess the racial and ethnic distribution of people stopped. Using the first traffic stop data reports from the San Jose Police Department as a case study, this paper examines baselines that are commonly used or discussed as appropriate. The paper argues that resident population data and/or official crime data are not adequate as baselines. As an alternative, the paper proposes an approach based on police early warning systems.