top of page

Evidence-Based Police Reform

Updated: Nov 12, 2020

There are 18,000 police departments and 800,000 officers in the United States. Departments vary in the quality of their leadership, professionalism, racial composition, and rates of officer misconduct. Misconduct may be limited to one or a few officers or the result of a widespread agency subculture that tolerates abuse. It is crucial, therefore, to avoid generalizing about “the police” and to instead target reforms toward departments where misconduct is systemic.

Reforms differ as well: some involve fundamental structural changes; some are more modest; and others can be considered minor. Some initiatives require a huge investment of resources. And some changes are guaranteed to meet with resistance from rank-and-file officers. It is easier to implement a reform than to sustain it over time.

The Black Lives Matter movement has called for the “defunding” of police departments. Defunding is not about abolishing police departments but, rather, is defined as shifting some law enforcement responsibilities to social and mental health services. Such reallocation can indeed help to address some of the root causes of crime, taking a more preventative rather than strictly reactive approach to crime. Positive results may be especially pronounced in disadvantaged neighborhoods, reducing both crime and the need for a large police presence in those neighborhoods. This is obviously a long-term process involving a host of stakeholders.


About 7 out of 10 Americans believe that police killings of citizens are a sign of “broader problems” in policing, rather than “isolated incidents.” The majority, from all racial groups, support a host of reforms. But abolishing or defunding the police are among the least popular.

  • Abolition vs. Defunding: 15% support “abolishing police departments,” but 47% endorse the idea of shifting some of the police budget to social programs.

  • Police Practices: Between 75% and 90% want more black officers hired, chokeholds banned, ending “stop & frisk” practices, mandatory body cameras, de-escalation training, and Early Intervention Systems to identify problem officers. 87% would require officers to report misconduct by their peers; 96% favor enhanced punishment for misconduct; and 78% support creating a commission of civilians to conduct oversight over the city’s police department. Remarkably, 56% want police unions abolished. (see Reuters Poll; NORC, University of Chicago Report; Yahoo News Poll; Gallup Poll).

  • Reforms: Other surveys document widespread support for demilitarizing departments, matching the racial composition of a department that of the city; creating civilian complaint review boards in cities that currently lack them; and robust community-policing practices.[1]


Some reforms have little chance of substantially improving policing.

  • Racial Diversification: Most studies indicate that diversifying a police department has little impact on officer behavior. Officers of different racial backgrounds tend to interact similarly with civilians, largely because of peer-group socialization and similarities in the challenges they face in their daily work. Many big-city police departments are majority black or Hispanic, but there is little evidence that their racial complexion correlates with improved police-community relations, less use of excessive force, or greater public approval of those departments (Pewtrusts, 2016). However, diversification does offer symbolic benefits: it shows that a department reflects the city’s population and is an equal-opportunity employer.

  • Training of new recruits usually makes little difference in police-citizen interactions, because officers are more sensitive to what their fellow officers expect (the police subculture) than what they are taught in the academy. However, serious in-service training (every one or two years) can help reduce officer misconduct, to some degree. Courses should prioritize use-of-force practices, de-escalation methods, and building positive relations with the community – in addition to firm messages about the penalties for violating departmental norms.[2]

  • Community Policing: Most police chiefs claim their departments are engaged in community policing, but the concept is vague and such efforts tend to be quite limited. In most departments, community policing is under the purview of a special Community Affairs unit and not a systemic philosophy and set of practices. Examples include occasional foot patrols, periodic police-community meetings, officer lectures in schools or involvement in youth programs, and so forth. These efforts tend to be either short-lived or marginal to the traditional enforcement orientation of most departments. It is rare for community-oriented practices to be embedded throughout the entire organization, which would include tangible incentives and rewards for officers who build healthy relationships with neighborhood residents. Departments involved in “problem-solving partnerships” with local residents have the advantage of identifying and remedying the underlying sources of recurring neighborhood crime problems, such as an abandoned building that is used to sell drugs.


Certain changes have the potential to improve police practices and enhance officer accountability. The most enlightened police chiefs in the country have already implemented them.

  • Body-cams can enhance officer accountability or prevent abuses in the first place (the former, but not latter, occurred in the George Floyd case in Minneapolis). The record is mixed as to whether they deter misconduct, but they clearly offer greater transparency, corroborating evidence, and enhanced accountability. Some studies have shown that use of body cameras reduces police use of force as well as citizen complaints against officers.[3]

  • Early Intervention Systems can be effective in identifying officers with a history of misconduct and correcting their behavior. Rule violations and citizen complaints are recorded in a database, and remedial measures (counseling, anger management, retraining, etc.) are imposed on officers who accumulate a threshold number of violations.[4]

  • Civilian Complaint Review Boards: Most large cities have such boards, which are responsible for adjudicating citizen complaints against specific officers. Complaints may involve excessive force, verbal abuse, corruption, and other misbehavior. There is no evidence that the presence of a review board deters officers from engaging in misconduct, but it can serve as one, among many, mechanisms of accountability, at least for the officers whose civilian complaints are upheld. But this means that the board or police chief must indeed impose punishment on rogue cops – something rarely done, as a recent database from New York City reveals. One way of ensuring that the complaint system operates properly is appointment of an Independent Auditor to monitor the process as well as related police practices, and periodically publish reports on the findings.[5] Civilian review boards deal with individual cases, whereas auditors have the advantage of addressing systemic issues. To date, only a few cities have such auditors.

  • Hiring Female Officers. The proportion of officers who are women has remained stable for decades: they comprise only 12% of American police departments (the proportion is significantly higher in Australia, Britain, and Canada). Yet research shows that female officers are less likely than their male colleagues to use physical force or draw their weapons, are less likely to injure suspects, receive far fewer citizen complaints, and are more able to defuse contentious encounters with members of the public. Regarding the latter, the 1992 Christopher Commission in Los Angeles concluded that female officers “feel less need to deal with defiance with immediate force or confrontational language.” It is surprising that hiring more female officers is rarely proposed as a way of reducing misconduct and building public trust in policing, but it clearly is something that should be embraced by reformers.

Other promising ideas – either proposed or already instituted in many departments – include requiring officers to record data on the persons they stop (to prevent racial profiling); use of independent prosecutors, outside the jurisdiction, to investigate fatal shootings by officers; and revising the scope of “qualified immunity,” the doctrine that shields officers from being sued personally for constitutional violations such as use of excessive force. If a lawsuit is successful, the money awarded to the plaintiff currently comes out of the city’s budget, not the accused officer’s bank account. Supreme Court Justice Sonia Sotomayor has criticized the immunity doctrine on the grounds that it “tells officers that they can shoot first and think later, and it tells the public that palpably unreasonable conduct will go unpunished.”

In the past, the Justice Department conducted investigations of police departments that showed a "pattern or practice" of misconduct or civil rights abuses. Agencies in about two dozen cities have been subject to comprehensive review and subsequently entered into consent decrees to ensure enactment of a package of reforms intended to transform the organization rather than piecemeal changes.[6] Although the measures agreed to are always challenging to implement and then sustain, they can yield dividends in improving police practices and public confidence. Moreover, the very fact that a department is being investigated by the federal government sends a message that fundamental remedies, not minor changes, are imperative, and that the top echelon of the U.S. Government is prepared to hold departments accountable.[7] The Trump administration ended this program in 2017. The capacity to catalyze both specific reforms and systemic, organizational change suggests that the Justice Department should resume these investigations.

Ronald Weitzer

Dept. of Sociology

George Washington University


[1] Ronald Weitzer & Steven Tuch, Race and Policing in America: Conflict and Reform. Cambridge University Press, 2006.

[2] Robin Engel, et al. “Does De-Escalation Training Work? Criminology & Public Policy, 19, 2020.

[3] Brett Chapman, “Body-Worn Cameras: What the Evidence Tells Us,” NIJ Journal, No. 280, January 2019. [4] Samuel Walker, Geoffrey Alpert, and Dennis Kenney, “Early Warning Systems,” National Institute of Justice, Research in Brief, July 2001. [5] Samuel Walker, Core Principles for an Effective Police Auditor’s Office, 2003. [6] Joshua Chanin, “Examining the Sustainability of Pattern or Practice Police Misconduct Reform,” Police Quarterly, 18, 2015; Amanda D’Souza et al., “Federal Investigations of Police Misconduct: A Multi-City Comparison,” Crime, Law, and Social Change, 71, 2019. [7] See also Final Report of the President’s Task Force on 21st Century Policing. Washington, DC, 2015.

Police Reform - final
Download PDF • 203KB

1,430 views1 comment

1 comentário

Thank you for a great overview of the issues. It's good to know that the public is far less polarized about the issue of police reform than the news leads us to believe.

bottom of page