Modified Census Benchmarking for Traffic Stop Statistics Analysis

Jack McDevitt, Northeastern University
Amy Farrell, Northeastern University
Shea W. Cronin, Northeastern University
Erica Pierce, Northeastern University

In response to allegations of racial profiling some state and local agencies have begun to track the race, ethnicity, and gender of those who are stopped and/or searched by police officers. By themselves, however, demographics of traffic stops are difficult to interpret. To remedy this problem researchers have begun to utilize external benchmarks such as census population data, driving population data, and road or violator surveys as a comparison for traffic stop data. Unfortunately, both residential population data based on the U.S. Census and driging population data gathered from road surveys have significant limitations. Residential population data may not always be an appropriate measure of the driving population in all jurisdictions. Although driving population data gathered from road surveys may more accurately represnt the driving demographics of a community such studies are often expensive, time consuming and subject to various interpretations. To remedy these weaknesses we have created a modified census population that weights the residential population data of a jurisdiction based on the demogrraphics and characteristics of surrounding communities. This paper presnts data from two communities in Rhode Island comparing the modified census benchmark results to other accepted benchmarks of driving demographics such as stationary observations.

(Return to Program Resources)

Updated 05/20/2006