U.S. Department of Justice


Date of this Version



Assessing Measurement Techniques for Identifying Race Ethnicity and Gender (January 2003) 8 pages


Over the past several years State and local governments have engaged in data collection regarding demographic characteristics of persons stopped by the police. These efforts are aimed at understanding factors used by law enforcement to make such stops. While data collections on law enforcement encounters have been undertaken, analysts have debated the availability of methods to meaningfully analyze these data. One complicating factor is the need to identify the baseline data necessary to make assessments regarding different racial groups’ experience with the police.

In 1999 several Federal law enforcement agencies designed and implemented data collection procedures to capture race and ethnicity information on persons stopped by their officers. The Immigration and Naturalization Service’s (INS) Border Patrol agents began collecting race and ethnicity data for those persons stopped at selected border crossings and highway checkpoints. Likewise Drug Enforcement Administration (DEA) agents collected data on nonspecific suspects stopped in selected airports.

At the State level, studies have been undertaken in Maryland, California, and New Jersey. New Jersey measured characteristics of persons using the New Jersey Turnpike and those that were speeding. (See list of sources.)

The various Federal and State agencies collecting the data wanted to have baseline data available to estimate the race and ethnicity of all persons passing through their area of responsibility. Lacking this kind of baseline data, it would not be possible to know whether the characteristics of the persons stopped were disproportionate to all those who had a probability of being stopped.


The Bureau of Justice Statistics (BJS), working with the Bureau of Transportation Statistics (BTS), contracted with Humanalysis, Inc., of Orlando, Florida, to conduct observational studies at two sites: 1) Immigration and Naturalization Service (INS) Border Patrol Checkpoint along Interstate 5 in San Clemente, California, and 2) the Detroit Metropolitan Airport. This report describes the findings from both data collections and assesses the difficulties in implementing this kind of a study.

The primary object of the observational studies was to determine the feasibility of using such techniques for estimating the demographic characteristics of persons coming through the checkpoint and airport, and what issues would be involved in replicating this technique in other locations. Can people’s race and ethnicity be easily recorded from observational techniques? To what extent do practical issues, such as gaining authorized access or proximity to persons under observation, play in the implementation of this kind of study?