Racial Bias in Traffic Searches

Purpose: We test current models of racial bias in policing, identify limitations, and propose a more comprehensive test of racial bias that accounts for population contraband rate. Methods: We conceptualize police officer search decisions as a 2 (search/no search) by 2 (contraband present/absent) table, with missing data (we do not know if contraband was present if the police did not search). Then we constrain the feasible problem space using properties of a 2x2 contingency table. Next, we compare all possible feasible 2x2 tables to identify instances of racial differences in police officer hit and false alarm rates. Results: For all feasible 2x2 tables we find that police officers have lower false alarm rates for White drivers than Black drivers. Conclusion:We interpret this difference in false alarm rate as a threshold bias (from Signal Detection Theory). That is, officers use different criteria for searching Black drivers than White drivers. We are limited by the available reported data from police departments. Future directions should explore how police officers make the decision to search drivers.}