NCJ Number
206651
Journal
Justice Research and Policy Volume: 6 Issue: 1 Dated: Spring 2004 Pages: 71-92
Date Published
2004
Length
22 pages
Annotation
This study compared police data on stops using the three most common benchmarks in racial profiling research: population estimates, field observations, and accident reports.
Abstract
The most important methodological concern in racial profiling research is the development of an appropriate benchmark to estimate the population of individuals available for police observation. Several methodological considerations can adversely affect the outcome of racial profiling studies, including the potential for reactivity in self-reported data and the accuracy of perceived racial and ethnic classifications. The purpose of this study was to compare the results of three independent racial profiling studies that were conducted in the same community, with the same police officers collecting data, and at the same time. The difference was in the type of benchmark used: population-based, field observations, and accident reports. The advantages and disadvantages of each benchmark are discussed, followed by a description of the study, which took place in Wichita from January to July 2001. Qualitative data on every official police/citizen contact were recorded by officers (n=37,454), including vehicle stops, pedestrian stops, bicycle stops, and traffic accident investigations. The analyses of this “Stop Study” data were performed using the three types of benchmarks. The comparison revealed only slight differences in the results; the noted differences in the conclusions of the studies seemed to lie in the a priori analytical assumptions and judgment of the researcher. Although important to research, benchmarks are of limited value to understanding the complex dynamics involved in racial profiling. However, the nature of research demands their use, thus it is imperative that researchers take the time to develop benchmarks that are accurate, defensible, and communicable. Tables, references