Strategies for Change

IN THE PRESS

Stanford News: "Stanford big data study finds racial disparities in Oakland, Calif., police behavior, offer solutions"

The Washington Post: "Oakland police, stopping and handcuffing disproportionate numbers of blacks, work to restore trust"

SF Gate: "Oakland police more likely to stop black people, study finds"

PBS News Hour: "Study slams troubled Oakland police department for racial bias"

Fast Company: "Stanford big data study finds racial disparity in Oakland police behavior"

The Christian Science Monitor: "What Oakland police's 'implicit bias' could mean for police reform"

Engadget: "Big data shows racial bias in police behavior"

Fusion: "Police talk more casually to black drivers than white drivers, a new study finds"

Futurity: "8 ways police can cut bias against African Americans"

EXECUTIVE SUMMARY

Law enforcement agencies across the United States are facing claims that they discriminate against community members of color. Inquiries into these claims often involve analyzing data from police stops. These so-called stop data reports typically take one of two approaches: either attack the agency for intentional racism, or deny the presence of racial disparities altogether. Yet neither of these approaches has yielded adequate progress toward many agencies’ mission of serving their communities with fairness and respect.

Taking a different approach, the City of Oakland engaged our team of Stanford social psychologists to examine relations between the Oakland Police Department (OPD) and the Oakland community, and then to develop evidence-based remedies for any racial disparities we might find. Racial disparities in policing likely have many causes. To examine these causes, our team has undertaken five research initiatives. We describe our research methods, findings, and recommendations in Strategies for Change: Research Initiatives and Recommendations To Improve Police-Community Relations in Oakland, Calif. We provide a technical report of our main research initiative, a thorough analysis of OPD stop data, in Data for Change: A Statistical Analysis of Police Stops, Searches, Handcuffings, and Arrests in Oakland, Calif., 2013-2014.

Across our research programs, we indeed uncovered evidence that OPD officers treat people of different races differently. At the same time, we found little evidence that these racial disparities arose from overt bias or purposeful discrimination. Instead, our research suggests that many subtle and unexamined cultural norms, beliefs, and practices sustain disparate outcomes. Our findings also suggest 50 evidence-based actions that agencies can take to change department cultures and strengthen police-community ties. Below, we highlight some of our research initiatives, findings, and recommendations for improving police-community relations in Oakland and other U.S. cities. Below, we highlight some of our research initiatives, findings, and recommendations for improving police-community relations in Oakland and other parts of the U.S.

The 5 Research Initiatives

• Statistical analyses of stop data from 28,119 forms that 510 OPD officers filed after stopping drivers and pedestrians in Oakland, Calif., between April 1, 2013 and April 30, 2014 (for a summary, see Chapter 1 of Strategies for Change; for the technical report, see Data for Change);
• Development of computational tools to analyze linguistic data from body-worn cameras (BWCs) and, using these tools, analyses of some 157,000 words spoken by OPD officers during 380 stops in April of 2014 (see Chapter 2 of Strategies for Change);
• Development of computational tools to analyze written narratives from police stop data forms, and, using these tools as well as human experts, analyses of some 1,000 OPD officer narratives from April of 2014 (see Chapter 3 of Strategies for Change);
• Two surveys of 416 Oakland community members regarding their attitudes toward and experiences with OPD officers (see Chapter 4 of Strategies for Change);
• Development and evaluation of implicit bias and procedural justice training modules with 675 OPD officers (see Chapter 5 of Strategies for Change).

Key Findings

• OPD officers stopped, searched, handcuffed, and arrested more African Americans than Whites, a finding that remained significant even after we controlled for neighborhood crime rates and demographics; officer race, gender, and experience; and other factors that shape police actions;
• Some 60% of OPD stops were of African Americans, who make up 28% of Oakland’s population;
• Of OPD officers making at least one stop during the 13-month period of study:
  • Only 20% stopped a White person, while 96% stopped an African American person;
  • Only 26% handcuffed a White person, while 72% handcuffed an African American person (excluding arrests);
  • Only 23% conducted a discretionary search of a White person, while 65% conducted a discretionary search of an African American person;
• When OPD officers could identify the community member’s race before a stop, they were much more likely to stop an African American, as compared to when officers could not identify the community member’s race;
• With African Americans, OPD officers used more severe legal language (e.g., mentioned probation, parole, and arrest) and offered fewer explanations for the stop than with Whites;
• In police-initiated interactions, African American and Hispanic Oakland residents felt more disrespected and misunderstood than did White and Asian Oakland residents.

Select Recommendations

• Our findings suggest the OPD has a culture where officers stop, search, handcuff, and arrest more African Americans than Whites. We suspect many other law enforcement agencies have similar cultures. In Strategies for Change, we thus recommend the OPD and other agencies regularly review their policies, practices, and procedures for evidence of disparate outcomes.
• As our findings reveal that less-experienced officers show more racial disparities in their stops, better training of new officers could likely reduce the degree of these disparities. To this end, Strategies for Change presents several recommendations for how to improve officer training.
• Although the OPD collects copious amounts of data, few measures track the OPD’s relationship with the community. In Strategies for Change, we thus recommend several actions that the OPD and other law enforcement agencies can take to measure what matters most.
• More broadly, we observe that many law enforcement agencies do not fully embrace data because they view it as evidence that could be used against them, rather than as feedback about what is or is not working, and why. In Strategies for Change, we recommend more than a dozen actions that the OPD and other law enforcement agencies can take to better leverage data.

Contact:

jleberhardt@stanford.edu

rhetey@stanford.edu