So what happens after someone is stopped for a non-moving violation? In some cases, the officer may search the driver, passengers or the vehicle. There are several ways an officer can search a car. One way is if he has probable cause - like seeing drugs openly in the car. Also if a person has an outstanding warrant, an officer is likely to conduct search.
Another way is that if an officer has a hunch, and suspects the motorist is involved in a crime, he may ask if he can conduct a search. This is known as a consent search. A driver has the choice to refuse the request, though criminologists say people rarely do.
For those concerned about racial profiling, including the National Organization of Black Law Enforcement Executives (NOBLE), an important question is whether the officer's suspicion proves to be right. Does he find contraband -- guns or drugs? The number of successful searches is known among criminologists as the "hit rate." So a key part of the question is whether the rate is similar for blacks and whites.
Among the 14 cities whose data were obtained by Dateline NBC, seven of them recorded searches, noted the type of search (including whether it was a probable cause search, a warrant-based search, or a consent search), and whether contraband was found.
In four cities - Houston, Denver, St. Louis and Nashville - after pulling drivers over for non-moving violations, there was no disparity between blacks and whites when it came to police finding guns or drugs.
But in three other cities - San Diego, Kansas City, Mo., and Minneapolis - consent searches of blacks were far less successful yielding guns or drugs. These searches, however, were not common. In these three cities, for example, less than 5 percent of equipment stops for either race led to a consent search.
- In Kansas City, which had 221,253 stops during 2001 and 2002, there were 435 consent searches following equipment stops for whites and blacks. Police found more guns and drugs, on average, on white motorists. Contraband was found 13 percent of the time on white motorists, and less than half as often - 6 percent - on vehicles driven by blacks.
- In San Diego, which had 121,013 stops in 2001, there were 251 consent searches following equipment stops for blacks and whites in which the department recorded whether contraband was found. Contraband was found 35 percent of the time on white motorists, and 26 percent - on vehicles driven by blacks.
- In Minneapolis, which had 60,166 stops for 2002, there were 67 consent searches following equipment stops for blacks and whites. These were searches in which the officer sought written consent from the driver. Contraband was found 26 percent of the time on white motorists, and less than half as often - 5 percent - on vehicles driven by blacks.
In response to questions by Dateline NBC, both Kansas City and San Diego's police departments cautioned against reading too much into these differences because consent searches following equipment stops happened so infrequently. None of the cities believed the statistics show racial profiling.
The Minneapolis Police Department has a policy against racial profiling. It said it is working with a committee that includes city staff, community members and law enforcement officials to draft a set of fairness standards and apply them as it analyzes both how police officers make decisions and the stop data it has collected.
Kenneth Novak, an assistant professor of criminal justice at the University of Missouri - Kansas City, who was hired as an outside consultant by the Kansas City Police Department to examine its data, wrote to the police chief after Dateline NBC asked about the difference in the hit rate that the department should monitor this gap.
But Novak also stated, "while this is an observation that deserves to be monitored within the department, it is still not clear to me what these estimates mean." He also wrote, "while these descriptive statistics may warrant further monitoring, clearly they do not represent routine or common events" in the department. He does not believe the numbers are statistically significant because they are a small number compared with the more than 200,000 stops during those two years. Novak wrote that he does not believe the consent search numbers show a racial bias.
San Diego Police Chief William M. Lansdowne wrote to Dateline NBC that because consent searches happen so infrequently during equipment stops, "it would be difficult to provide reasonable explanations for the differences." He also wrote that "Some researchers argue that these consent search rates are neither unfair nor unreasonable given the over representation of Blacks/African Americans among described crime suspects."