Skip to main content Skip to secondary navigation

We work to understand the impact of existing criminal justice policies and practices, and apply data-driven approaches to make the criminal justice system more equitable.

Criminal Justice

Main content start

Understanding the Impact of Existing Law Enforcement Practices on Communities

Law enforcement has been at the center of conversations about race in the U.S. for decades, and especially in the last few years. SPARQ leverages big data to quantify the impact of existing law enforcement practices on communities and identify actionable ways to mitigate racial disparities.

Leveraging Stop Data

police car parked at night

Law enforcement disproportionately stops Black drivers for routine traffic stops. These kinds of disparities erode communities’ trust in police. SPARQ partnered with the City of Oakland and the Oakland Police Department, and analyzed the data OPD routinely collects about traffic stops. The analysis found stark racial disparities in who was stopped, searched, handcuffed, and arrested. Results and recommendations are reported in Data for Change and Strategies for Change.

Developing Tools to Analyze Body-Worn Camera Footage

Police body-worn camera

Communities of color have long described mistreatment by police during routine encounters. Working with the Oakland Police Department and the Natural Language Processing Lab, SPARQ developed a method to transform officers’ body-worn camera footage into big data about police-community interactions. This work showed that officers use less respectful language and tone of voice when addressing Black drivers, and that traffic stops have routinized conversational structures. SPARQ is working to develop and evaluate police trainings and other reforms to reduce these disparities. Read more about our research on respectful language, tone of voice, and conversational sequence of traffic stops.

Exploring the Language of Escalation

A police officer talking to a man during a traffic stop

In the wake of the murder of George Floyd, more jurisdictions have begun to restrict officers from conducting routine stops to reduce policing’s disproportionate impact on communities of color. When police officers interact with community members, it is critical they avoid unnecessary escalation and prevent harm. With the Natural Language Processing Lab, SPARQ is exploring  the linguistic signature of escalation in routine traffic stops using body-camera footage shared by a partnering agency. For stops of Black drivers, initial findings show that the words officers spoke in the first 30 seconds predicted how the stop would end. SPARQ is working with police departments to incorporate these findings into de-escalation training.

Applying Data-Driven Approaches to Reimagining Public Safety and Reducing Disparities in the Criminal Justice System

With so much urgency surrounding calls to reimagine public safety and reduce racial disparities in the criminal justice system, SPARQ is working to chart an evidence-based path forward.

Building a Network of Law Enforcement Agencies

Police officers in a meeting

The national spotlight on policing has renewed the focus on inequities and challenges facing the profession. Documenting and better understanding these problems can pave the way for identifying solutions. In 2018, SPARQ began building a network of law enforcement agencies in California to co-develop and test data-centered solutions to build trust with the community and make policing more equitable.

 Innovating Police Trainings

Jennifer Eberhardt and OPD Chief LeRonne Armstrong

With the push to reimagine policing and public safety came an increased demand for officer training, yet relatively little is known about their effectiveness in shifting the culture of law enforcement. SPARQ is currently partnering with police departments to study, develop, implement, and evaluate trainings designed to innovate police practice, strengthen relations with the community, and improve the culture of policing. To learn more about our initial research on police officer training, see our white papers: Principled Policing: Procedural Justice and Implicit Bias Training and Principled Policing: A Path to Building Better Police-Community Relations.

Measuring the Impact of Policy Change

It is hard to change institutions and social systems. With the Oakland and San Francisco Police Departments, SPARQ is applying a data-driven approach to evaluate the impact of policy reform within law enforcement. SPARQ is developing a method to evaluate how policy change directly translates into officers’ actions and equitable outcomes for community members.

Understanding the Role of Race in Criminal Justice Decision-Making


Law enforcement is the entry point to the criminal justice system. Police set in motion a process that can go on to prosecution, adjudication, sentencing, and ultimately corrections. It is important to understand the role of race in decision-making more broadly and across a variety of institutional actors within the criminal justice system (e.g., judges). In collaboration with the Vera Institute of Justice, SPARQ is exploring how race and other factors impact the decision-making of institutional actors.

Analyzing Narratives about Crime and Criminal Justice Reform

crime scene tape

Reforming the criminal justice system does not happen in a vacuum. Stakeholders are constantly exposed to broader narratives about race and crime, the purpose of the criminal justice system, and what the consequences of altering the system might be. SPARQ is exploring the shifting landscape of public safety and criminal justice reform and how exposure to different messages might affect public attitudes and policy preferences.