Research Highlights

Do You Trust the Police?

How do you ask people in gang-ruled neighborhood whether they trust the police? A team of researchers are using a Matrix seminar to explore the most effective approaches for measuring communities' attitudes toward law enforcement.


For policy-makers and law enforcement officials working to address crime and insecurity, establishing trust-based relationships in local communities is essential. Yet it has traditionally been challenging to get an accurate measure of trust, especially when gang members and other criminals may threaten retaliation for cooperation with police.

To tackle this challenge, UC Berkeley Social Science Matrix is supporting a team of researchers from diverse social science disciplines to explore new approaches of measuring perceptions of law enforcement in local communities. Bringing together methods from fields such as political science, psychology, and development economics, the group hopes to examine how “sensitive survey design” methods could more accurately measure attitudes toward law enforcement in communities with organized criminal gangs.

The goal is “to understand how police departments can transform relationships with local communities in cases where community trust and confidence in the police is low,” explains the seminar’s co-leader, Aila Matanock, Assistant Professor of Political Science at UC Berkeley, who says she hopes the seminar will lead to a “cross-fertilization of ideas and methods” and “building a larger community of people who are interested in these questions moving forward.”

If people are more supportive, they’re more likely to support investigatory practices, and if they’re skeptical or cynical, they’re less likely to report crimes.

Matanock first saw the value of experimental survey design in her research measuring public perceptions of the military in Colombia. “In the survey, some respondents were asked direct questions, like ‘Would you support the military having more independence in the counterinsurgency,’ while others were asked indirect questions, like ‘How many of the items in this list do you support?’… The two methods showed very different attitudes.”

Matanock is coordinating the seminar with Jack Glaser, Associate Professor and Associate Dean at UC Berkeley’s Goldman School of Public Policy, who has worked with the Department of Justice and other institutions to address racial profiling. “In some places, norms can dictate that it’s good to support the police, and in other places, norms can dictate that people should be critical of the police,” Glaser says. “You’re not going to get an accurate read if you ask the questions too directly. It’s going to be hard to predict.”

Through the Matrix seminar, the two researchers hope to bring new methods to a local context, by working with local law enforcement officials from Oakland to assess community trust in police. “We’re thinking about how to integrate it into the U.S. context, building off a larger school of literature that’s applying social psychology techniques to political science,” Matanock says. “These are questions facing Oakland and other police departments around the Bay Area.”

Like all Social Science Matrix initiatives, this seminar is cross-disciplinary by design. In addition to considering surveying, the seminar will also consider how increased use of new surveillance technology and intervention by external actors—especially federal oversight, consultants, and private security guards—can influence community trust in the police.

“Being able to tap those community attitudes is really important,” Glaser says. “It’s important to police chiefs; they really care about community attitudes. It’s an area where scholars, policy people, the practitioners, and the community can all get together and work toward the same goal…. If people are more supportive, they’re more likely to support investigatory practices, and if they’re skeptical or cynical, they’re less likely to report crimes. It matters on the ground.”

Photo Credit: Chris Huggins.

You May Like


Learn More >


Learn More >


Learn More >