Analyzing police stops

Illustration of a police traffic stop at night.

In collaboration with Big Local News, the Computational Policy Lab, and the Policing Project.

Police stops are the most common way that the American public interacts with law enforcement. But policymakers lacked tools to answer important questions about police stop practices, including tools to measure their efficacy and impact.

We developed new approaches to help policymakers answer these questions. Our analyses opened the door to major policy changes—like a 70% drop in ineffective traffic stops in Nashville—and a state requirement in California mandating transparent reporting of police stop activity by law enforcement agencies statewide.

Noting the lack of comprehensive national data on police stops, we worked with colleagues to assemble, standardize, and publish the largest public archive of police stop data in the U.S., comprising over 250 million stops from 33 state patrols and 56 local police departments. Our rigorous statistical analysis of this data provided one of the most comprehensive demonstrations of racial disparities, enabling policymakers and law enforcement agencies to benchmark practices and evaluate the effectiveness of new interventions aimed at reducing bias.

As part of this effort, we published free statistical tutorials so anyone could replicate our analysis of policing practices. Our dataset now serves as a scholarly and educational resource, having been cited in over 700 studies and used in courses across the country. We also collaborated with investigative journalists from major outlets like the LA Times, who used the data we assembled to conduct detailed analyses, leading to local reporting that spurred policy changes in numerous jurisdictions.

Learn more at openpolicing.stanford.edu and in our peer-reviewed paper, “A large-scale analysis of racial disparities in police stops across the United States.”

Contributors