SOME EVIDENCE ON STOPS AND SEARCHES

We test derivations from models of statistical discrimination and preferential discrimination with observational data on police-suspect interactions in two American cities. In neither city do we find evidence of preferential discrimination against African American suspects. In both cities, using a preferential model of searches following suspect stops, we find evidence of preference extended to white suspects by African American police officers. We offer reasons why we did not find the frequently alleged pattern of racial profiling, and speculate on our otherwise unreported finding for white suspects In laypersons ’ parlance, police racial profiling occurs when police use race to determine which citizens are subjected to police authority, such as stops, searches, and various forms of coercion. For more than a decade, American police, politicians, and the public have been concerned about the problem of racial profiling, giving the issue high visibility. Most of the American public seems convinced that it occurs (Gallup Poll News Service, 1999; Penn, Schoen and Berland Associates, 2000), but debates are unresolved about where it occurs, how severe it is, and whether and when it is a problem that reflects unjustified bias. A recent review of research on this topic concluded that insights into the nature and extent of this phenomeno