Recent events have put police in a bind. A surge of violence means cops are needed more than ever, something now acknowledged across the political spectrum. But officers’ behavior remains under a microscope, between the spread of cell phones and body cameras, the waves of anti-police sentiment that follow high-profile incidents such as the killing of George Floyd, and the conflicted attitudes that many Americans have toward cops. Few Americans of any demographic group actually want less policing; yet stark racial differences exist in how much faith Americans have in cops to do their jobs fairly.

Ideally, we’d be able to put more officers where they’re needed most while minimizing unnecessary tensions between cops and the public. A new study from George Mason University criminologist David Weisburd and several coauthors suggests that “procedural justice” training might be a way to achieve that.

“Procedural justice” is a gimmicky academic term for a straightforward idea: people are more comfortable with authority when they feel like they’re being treated fairly and respectfully and get a chance to voice their concerns. It’s become increasingly common to teach officers that these principles are important—that cops should, for example, be courteous, demonstrate that they have trustworthy motives, and listen attentively when civilians express their views—and to have cops practice them in role-playing exercises. Research into the effects of this training has been promising but sparse.

The strength of the new study is that it’s a randomized experiment conducted in three cities. Participating cops were randomly assigned either to receive 40 hours of procedural-justice (PJ) training or not, and then deployed to cover high-crime “hot spots” for a nine-month period. (Both groups were briefly trained in hot-spot policing and taught how to record data for the project, and the PJ group later received a refresher course, too.) The experiment started in July 2017 in Tucson, in February 2019 in Cambridge, and in October 2019 in Houston—where the pandemic took hold before the project was completed.

The researchers studied what happened next, not just by monitoring crime trends at these locations, but also by surveying community residents on their opinions about cops and even sending investigators on ride-alongs with the officers to assess their treatment of civilians (“blinded” to whether the officers had received the PJ training). In short, the study was a remarkable undertaking, combining rigorous scientific design with labor-intensive execution.

The approach has limitations, though. First, only 28 officers were involved across all three cities, a rather small sample—and these cops provided only about half of the policing of the “hot spots” to which they were assigned. Second, these cops knew they were being studied, not least because they filled out activity logs on forms that reminded them about the principles of procedural justice; this could have made them follow the training more closely than they otherwise might have. And third, to be included in the study, officers had to volunteer, and the departments were told to pick cops who “had at least a few years of experience, were unlikely to be promoted during the intervention, and had a reputation for being active on patrol”—meaning the results might not generalize to the broader population of officers.

Nonetheless, the results were striking. Local residents became less likely to report that the police were using too much force or “harassing” civilians, the evaluators found that the trained officers followed the principles of PJ more than the untrained ones, and the trained officers made far fewer arrests—about 60 percent fewer.

These cops maintained a presence in crime hot spots while creating less friction with residents, which is a good thing. Still, one might worry that the decline in arrests reflects softness rather than a smart, diplomatic approach. A key question, of course, is what effect the training had on crime.

The authors measure crime in two ways. The hot spots worked by the PJ-trained officers saw 14 percent fewer recorded “crime incidents,” though this could partly be a result of the drop in arrests. (Some incidents are documented via cops’ own enforcement behavior, though the authors note that Cambridge, where cops made few arrests in general, saw the same effect in the areas handled by PJ-trained cops.) Citizen-initiated calls—which could also have been affected by cops’ behavior, as manipulated by the experiment—show an effect in the same direction, though it is too small to be considered statistically significant.

The crime results are somewhat uncertain but produce no evidence that reducing arrests in this way increased crime. If anything, the PJ approach to policing may help reduce it. The study leaves some questions unanswered, and departments relying on it should carefully monitor the effects of their own efforts. But it suggests that procedural-justice training could be the key to boosting police presence in places that need it without stoking community backlash.

Photo by Loren Elliott/Getty Images


City Journal is a publication of the Manhattan Institute for Policy Research (MI), a leading free-market think tank. Are you interested in supporting the magazine? As a 501(c)(3) nonprofit, donations in support of MI and City Journal are fully tax-deductible as provided by law (EIN #13-2912529).

Further Reading

Up Next