Los Angeles, CA
Everyday Respect: Measuring & Improving Police Officer Communication During Motor Vehicle Stops
Organization: University of Southern California
Primary Investigator: Morteza Dehghani
Research Track: Resource and Service Equity
Traffic stops are both mundane and potentially deadly. Past studies have shown harmful racial disparities in how civilians are communicated with during these stops. The Everyday Respect project uses body-worn video recordings to study the interaction of different forms of communication and contextual factors by Los Angeles Police Department (LAPD) officers and drivers during traffic stops. We assess how different stakeholders view “good” officer communication; build machine learning tools to measure communication; and then work with LAPD to design training and policy innovations based on our analysis. Even though we focus on LAPD communications, the machine learning frameworks, the officer training curricula, and the policy innovations we develop are potentially applicable nationally.
NSF Abstract
Traffic stops carried out by law enforcement officers include routine encounters as well as situations that hold the potential for fatalities. This project aims to investigate the communication dynamics during traffic stops conducted using audio and video footage from body-worn cameras. The research builds on evidence that officers' initial communication influences the course of the interaction and whether it escalates or not. Additionally, the project will examine potential disparities in the treatment of community members based on demographic factors and other relevant variables. The project will also study the interplay among factors such as demographic characteristics, disability status, community context, and officer training in shaping officer and driver interactions. Collaboration with a training academy is integral to developing and implementing new training curricula based on the research findings.
This study takes a community-informed approach by incorporating footage from body-worn cameras, as well as complementary data on stops, personnel, driver behavior, and community context. The project includes input from the department studied and other community stakeholders to define the dimensions of effective communication that should be evaluated. Human annotators from diverse backgrounds are employed to code the identified dimensions of officer communication. Machine learning tools, trained on these human annotations, enable the project team to analyze communication patterns at scale. The project team conducts statistical analyses to investigate the causes and consequences of officer communication, including factors that may contribute to differential treatment of different groups within the community, and to develop new training tools that can be tested at the training academy.
The CIVIC Innovation Challenge is a collaboration with Department of Energy, Department of Homeland Security, and the National Science Foundation.
This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.