In the controversy over racial profiling, reformers have generally advocated that police departments collect data on traffic stops. Several states have laws requiring data collection, and a bill requiring it has been before Congress for many years.
Almost all current traffic stop data efforts, however, involve aggregate data (e.g., the percentage of stops that involve African Americans), and use the resident population as the base line.
There are many limitations with this approach, however. Read Sam Walker’s article on this issue, “Searching for the Denominator” in the Journal of Research and Policy (2001).
The best book on the subject is Lorie A. Fridell, By the Numbers: A Guide for Analyzing Race Data from Vehicle Stops (PERF, 2004) for the best discussion of all the different methods for analyzing traffic stop data.
The problem with aggregate data is that they do not identify individual officers. And so it is not possible to identify particular officers, or shifts, or department units where there might be a problem.
A solution to this problem is to utilize an Early Intervention System (EIS). Go to the EIS page for a discussion of EIS.
An EIS allows a process of Internal Benchmarking that can effectively identify particular officers who may be stopping and searching a disparate number of people of color. See Sam Walker’s paper on Internal Benchmarking. Internal benchmarking is discussed in detail in Chapter 8 of Fridell, By the Numbers.