The Washington PostDemocracy Dies in Darkness

Guest Post: Data can transform policing — but first we need the data

A call for law enforcement to provide the information that better directs the police reform movement

In the ongoing discussions about police reform, one aspect that is frequently cited but rarely fixed is the lack of information from law enforcement agencies on what they do and how often they use force. Here, noted criminologist Nancy La Vigne and former top Justice Department official Roy L. Austin Jr. discuss the importance of getting that data to fully inform the process of reform:

As calls for police reform continue, a deceptively mundane issue — the glaring lack of data on police activities and outcomes — is failing to receive the attention it deserves.

We cannot fix a problem we do not fully understand. And we do not fully understand the problem because the available data on policing is pathetically thin. This is inexcusable, and it is a serious obstacle to progress.

For a second year, most U.S. police departments decline to share information on their use of force

From our recent work on the Council on Criminal Justice Task Force on Policing, we are well-schooled on the woeful lack of good data on law enforcement and acutely aware of how democratizing such data could benefit everyone. Here are four examples:

Use of Force

As documented in the task force report, “Policing by the Numbers,” confirmed cases of people killed by law enforcement are not systematically tallied and lack contextual details. As such, it is impossible to know which police killings involve armed or unarmed citizens, “suicide by cop” episodes, domestic situations or active shooters. Even short of fatalities, we have no comprehensive source of police use-of-force data in this country.

In 2019, the FBI established the National Use of Force Data Collection. But the program is strictly voluntary, and just 27 percent of agencies have contributed data so far. According to incomprehensible FBI rules, most of these data may not be accessed until at least 81 percent of officers are represented in the database. Even at that point, the data will only be released in the aggregate, meaning that the public will not know which individual agencies have unusually high use-of-force rates and should be held to account.

Police Misconduct

Data on police misconduct is uneven at best. Officers who engage in wrongdoing may resign before termination with cause, and even those who are fired often find employment elsewhere. This is troubling for many reasons, and especially because we know police officers who misuse force once are more likely to do so again.

Police decertification databases exist in most states to prevent this outcome, but they are far from comprehensive. A national police misconduct registry such as that provided for in the George Floyd Justice in Policing Act (H.R. 1280) would ensure documentation of misconduct cases.

Racially Disparate Policing

Questions around racially disparate policing have played a central role in the public’s resounding demand for more equitable law enforcement practices. But we have scant empirical data on the nature and degree of these disparities. When access to such data is available, the findings are sobering. Take traffic stops, for instance. One study of 35 police departments found that Black drivers were almost twice as likely to be stopped as White drivers. This difference shrank somewhat for stops conducted after sunset, when it is more difficult for an officer to discern the driver’s race.

Documenting such biases in policing can contribute to reform in many ways. For researchers, it provides a critical baseline against which we can assess the impact of changes to officer training and other measures designed to promote more equitable policing.

Patrol Activity

We have very little notion of how officers spend their time. Such data are essential to assess the opportunities and trade-offs associated with shifting police functions to other actors or organizations. For example, an analysis of dispatch data from nine jurisdictions found that traffic-related calls occupied about 18 percent of officer response time, yet mental health-related issues made up a very small share of calls and consumed only 2.2 percent of officer time.

Before adjusting police responsibilities, cities should analyze the share of calls and encounters associated with those services targeted for offloading. Information on the time officers spend responding to calls, investigating crimes and engaging in other activities is essential to guide funding decisions.

There is one bit of good news about the lack of data transparency in policing: The solution is relatively simple. The Biden administration could spur greater access to data by offering federal grant incentives and requiring data collection and transparency as a condition of agencies’ receiving such funds. Data dashboards and scorecards, such as the Police Data Initiative, that enable interactive queries of the data in real time, would further democratize these data and should be commonplace.

Overall, we must remember that police data do not belong to police — they belong to the communities that give police their authority. While data may not be the most inspiring topic in the national conversation about police reform, its importance in the quest to promote more equitable law enforcement cannot be overstated.

Roy L. Austin, Jr. is a former deputy assistant attorney general in the Civil Rights Division and a member of the Council on Criminal Justice Task Force on Policing. Nancy La Vigne, a criminologist with expertise in a wide array of criminal justice reform measures, is executive director of the Council on Criminal Justice Task Force on Policing.