Changing How Police View the Public: Lessons from the Philippines

Around the world, community policing is often promoted as a way to repair strained relations between officers and citizens. The idea is straightforward: If police spend more time with residents in informal settings, they will better understand local concerns and see citizens as partners rather than adversaries.

Although this idea has intuitive appeal, almost all research focuses on how engagement between police officers and community members (through nonenforcement settings like conversations with local leaders or household visits) affects citizens. We know very little about how it affects officers themselves. This is a major gap because efforts to improve police–community relations require changes on both sides. If officers see citizens as adversaries rather than partners, they may continue to neglect or abuse citizens despite community policing, and community policing will fail to deliver on its promises.

Our study evaluates this assumption directly. Working with the Philippine National Police in Sorsogon Province, we implemented a large-scale randomized experiment to test whether sustained contact with civilians changes officer attitudes. The results challenge some of the more optimistic claims surrounding community policing, but they also point to the conditions under which these programs are more likely to work. Together, the findings help explain both why community policing often falls short and where it has real potential.

The Puzzle

Community policing is built on a specific logic. Routine, face-to-face engagement with residents is supposed to correct officers’ misperceptions about the communities they serve, reduce suspicion, and increase attention to citizen well-being. Yet we have little experimental evidence about whether these changes actually occur. For officers, daily work often involves responding to crime, dealing with disputes, and interacting with people during moments of stress or fear. These encounters can reinforce a negative view of the public based on narrow interactions with just a subset of the population. The question is whether structured, positive contact with people in the community can shift that mindset. This became the central puzzle of our research.

The Experiment

Sorsogon is a diverse province of roughly one million residents spread across semi-urban centers, coastal towns, lowland villages, and upland rural areas. Before the intervention, nearly half of surveyed residents saw a police officer once a month or less. To increase engagement, we partnered with local police leadership to assign a random subset of the province’s 705 frontline officers to intensive community-policing activities for seven months. Officers conducted household visits, met with barangay leaders, and engaged residents in a range of nonenforcement interactions. Because the assignment was random, we could cleanly identify whether increased contact caused changes in officer attitudes.

What We Found

The results reveal a mixed picture. Officers gained a better understanding of citizen concerns. This is important because it corrects the common misperception that residents care only about high-profile or violent crimes. Officers learned that issues like neighborhood disputes, minor theft, and public intoxication often matter far more to citizens than they initially believed. This shift in understanding moves officers closer to the concerns of the community, one of the central goals of community policing.

Other attitudes, however, did not change. Officers did not develop greater empathy for civilians, nor did they become more trusting of residents. They also did not express stronger beliefs that misconduct is serious or that citizens deserve greater protection from abuse. These outcomes are central to the philosophy behind community policing, yet officer attitudes proved resistant to change. The findings suggest that while contact helps officers learn what matters to residents, it does not automatically transform their deeper attitudes about accountability, citizen motivation, or the seriousness of misconduct.

A More Nuanced Theory of Contact

To understand why these effects were uneven, we complemented the experiment with qualitative observations and exploratory analyses. The emerging explanation is that community contact improved attitudes among officers who were not already embedded in the communities where they work. For these officers, the additional engagement provided new information and a more accurate picture of residents’ concerns. On the other hand, the intervention offered little new insight for officers who already lived in or near their assigned communities. Contact did not expand their understanding because they already had preexisting ties.

In some cases, contact even produced negative reactions. Officers who perceived personal safety threats became more wary of community members after engagement. This pattern was especially evident in areas with higher risks of violence. When officers interpreted contact as exposing them to new dangers rather than building understanding, the intervention strengthened suspicion instead of reducing it. This dynamic helps explain why community policing programs sometimes fail in other contexts. Contact itself is not universally reassuring. It can either reduce or heighten perceived threat depending on what officers encounter.

Implications

These findings offer a more realistic view of what community policing can accomplish. Community policing can help officers understand the public’s concerns more accurately, which is a meaningful step toward better service delivery. However, contact alone is not enough to produce broader changes in empathy, trust, or accountability. These shifts likely require stronger institutional incentives, clearer norms regarding misconduct, and organizational structures that reinforce collaboration and transparency.

One reason contact alone does not produce shifts in empathy or accountability is that officers still operate in environments where misconduct carries few real consequences and where rewards are tied mainly to enforcement outputs. Changing that dynamic requires going beyond community engagement. Two kinds of reforms are especially promising. First, independent oversight or internal disciplinary systems with clear processes and transparent consequences can strengthen norms around professionalism and reduce tolerance for abuse. Second, promotion and evaluation criteria that reward problem-solving and community cooperation, rather than focusing only on arrests or case clearance, can give officers real incentives to build relationships and respond to citizen concerns. These structural changes can help support and sustain the attitudinal improvements that community policing alone cannot deliver.

Beyond these structural changes, the effects of community policing also depend heavily on officers’ starting points. Officers who lack local ties appear to benefit the most from structured contact. In contrast, officers who already have community connections or who feel vulnerable may not respond as reformers expect. The lesson is that community policing is not a universal fix. It is a tool that can help under certain conditions and among certain officers, but it cannot substitute for deeper reforms that address organizational culture and structural incentives.

The Broader Lesson

Many governments continue to invest in community policing because it promises to humanize interactions and build trust: Brazil’s Pacifying Police Units were created to build trust in Rio’s favelas. South Africa’s Community Policing Forums were designed to give residents a voice in local safety decisions. Indonesia’s Bhabinkamtibmas program assigns officers to villages to maintain daily contact with community members. These efforts all reflect the same hope that closer engagement will change police-community relations, but our findings suggest that deeper internal reforms are needed for these programs to reach their full potential. Community contact can correct officers’ misperceptions about citizen priorities, but it does not automatically transform the attitudes most linked to abuse and discrimination. Reformers should view community policing as one part of a larger strategy to improve police-citizen relations. It can enhance understanding, but it cannot solve structural problems on its own.

Dotan Haim is an assistant professor of political science at Florida State University. Matthew Nanes is associate director and associate senior instructional professor at the University of Chicago’s Committee on International Relations. Nico Ravanilla is an assistant professor of political science at the UC San Diego School of Global Policy & Strategy.

Thumbnail credit: Mico Medel (Pexels)