Following the militarized police response to protestors in Ferguson, MO, precipitated by the police shooting of Michael Brown, President Obama’s Task Force on 21st Century Policing was convened to recommend reforms to a policing profession experiencing a crisis of trust and legitimacy. The Task Force identified building trust and legitimacy as the foundational principle underlying just and effective policing. A key policy recommendation was to allocate more resources toward community-oriented policing practices aimed at increasing trust and cooperation through positive, non-enforcement interactions between police officers and the communities they serve.

In contrast to policing that emphasizes punitive enforcement, community policing encourages the formation of cooperative relations through a variety of non-enforcement interactions. Over the decades, community policing has evolved haphazardly across the U.S. to encompass an unstandardized patchwork of interventions that includes walking patrols, police-led sports leagues, and community meetings. Proponents of this policing paradigm maintain that the positive interactions at the heart of these interventions are a necessary step towards repairing the broken relationships between police and the public, particularly in minority communities where police-community relations are often characterized by longstanding conflict and distrust.

Despite its popularity and dedicated funding streams, the core assumption behind community policing — that positive, non-enforcement interaction between police and the public can improve attitudes toward police — has proven difficult to evaluate with evidence. A 2018 report by the National Academy of Sciences, for example, characterized the few empirical studies in this domain as insufficient, concluding that “research is needed that tests the ability of a single interaction to shape general views about police legitimacy”.

The Community Policing and Police Legitimacy project emerged from a collaboration between Kyle Peyton (Yale), Michael Sierra-Arévalo (Rutgers), David Rand (MIT) the New Haven Police Department (NHPD) to test whether positive, non-enforcement interactions with police officers actually cause meaningful improvements to police-community relations. This project, a joint initiative between the Justice Collaboratory and the Institution for Social and Policy Studies, combines qualitative fieldwork with randomized experiments and computational social science to study how interactions between police and those they serve shape the attitudes that underlie the cooperative relationships necessary for public safety. Led by Peyton, the team recently conducted a first-of-its-kind randomized field experiment — a real-world version of a clinical drug trial — to test whether positive, non-enforcement interactions between police and the public can actually change how individuals view the police.

First, they conducted a baseline survey of New Haven residents to measure their attitudes toward police. Then, half of those who responded were randomly assigned to receive a community policing “treatment”, a short, non-enforcement interaction with a New Haven police officer who knocked on the door, introduced themselves, solicited feedback about neighborhood issues, and provided residents with their department issued cell phone number. The other half of respondents were assigned to a “control” group that did not receive a community policing visit.  After the community policing visits, residents in both the treatment and control groups were re-surveyed twice — three days after the community policing visits and again at 21 days. This design allowed the team to test not only if a single community policing visit caused a change in attitudes toward police, but also whether the effects of a visit lasted more than a few days.

The team found that a single positive, nonenforcement interaction with a police officer significantly improved police legitimacy, perceptions of police performance, and willingness to cooperate with police to solve crime and neighborhood problems. These effects persisted for up to 21 days after the community policing visits. Moreover, they found these effects were not driven by improved attitudes among residents who already had positive views of police. Instead, the largest improvements were among Black residents and those who felt most negatively toward police before the community policing visit. Among Black residents, nonenforcement contact with an officer had an initial effect that was nearly twice as large as the effect observed among White residents, a notable finding given the longstanding tensions between Black communities and police in the United States.

This study, published in the Proceedings of the National Academy of Sciences, provides the first causal evidence that positive, non-enforcement interactions between police and those they serve can shape the attitudes that underlie the cooperative relationships necessary for public safety. As the evidence of punishment-centric policing’s costs to community wellbeing continues to mount, the research suggests that community policing like that tested in New Haven can be a valuable tool to build trust and promote cooperative relationships between the police and the public. In lieu of single-minded reliance on stops, searches, and arrests, police leaders can benefit from investment in infrastructure to record and reward community policing interactions that are often only supported in theory. Insofar as police departments encourage their officers to interact with the public as collaborators in a shared mission of community wellbeing, it holds tremendous promise.


Kyle Peyton
Michael Sierra-Arévalo
David Rand
New Haven Police Department