Search About Newsletters Donate
We’re celebrating our 10th birthday!

Keep a bright light shining on the darkest corners of the criminal justice system. Become a member of The Marshall Project today.

A police cadet works on a computer in the 1980s. Despite rapid advances in technology, few departments are capable or even willing to collect and share important data.
Analysis

13 Important Questions About Criminal Justice We Can’t Answer

And the government can’t either.

A few weeks ago, the White House trumpeted the progress of its Police Data Initiative. The nearly one-year-old project prods local cops to publish data on their operations in a bid to increase transparency and build trust with the communities they police.

The results were underwhelming. Of nearly 18,000 police agencies from coast to coast, just 53 had signed on to the effort. Of that inaugural class, eight released data on officer-involved shootings, and six published information on their officers’ use of force.

After the deaths of Freddie Gray and Laquan McDonald and others—in an age when police in many cities are under greater scrutiny than they’ve been in decades—how is it that we know so little about how officers employ force to subdue suspects?

“How many times this week has the department used a Taser? How many times have people been hospitalized because of a beating?” said Walter Katz, the independent police auditor for the city of San Jose, Calif. “There is a complete dearth of such information. To me, that type of force can have just as corrosive an effect on community relations as an officer-involved shooting.”

The open secret is that we know very little about much of how the criminal justice system operates in America. These aren’t things the government knows and won’t tell us (though there are plenty of those, too). It’s because state, local and federal governments, which ought to rely on data to inform the policies they enact, just don’t know.

In some cases, the federal government commissions criminal justice surveys that offer national estimates, often years after the fact. But the kind of granular, local, real-time data that powers most industries is all but absent. The number of times police use force or shoot someone in the line of duty are just the most obvious examples in our current national conversation.

Among the things we don’t know about our criminal justice system:

The excuses for why we don’t have better data about our police, our courts and our prisons may sound familiar to anyone who has worked in corporate America: there isn’t enough money to hire analysts; the IT department says it can’t be done; the chief is moving on to another department.

Local autonomy has not been helpful for good criminal justice data. The fraction of the country’s 18,000 police departments that do collect figures on officers’ use of force have no consistent definition of what constitutes force. Adam Gelb, director of the Pew Charitable Trusts’ Public Safety Performance Project, cites similar issues in other parts of the system, like probation. There are thousands of probation agencies, but they are either run at the state or local level. In one place, probation is part of the executive branch; In another, it’s part of the judiciary. The lack of consistency makes contacting all the agencies a daunting prospect, much less moving them toward timely and uniform reporting of statistics.

The Center for Policing Equity has been collecting data from police agencies on pedestrian and traffic stops as well as uses of force. The center’s co-founder Phillip Atiba Goff, a visiting scholar at Harvard’s Kennedy School of Government, said the National Justice Database has commitments from police departments that cover about one-third of the nation’s population. But for him, it’s not a matter of whether police are collecting data, but how they’re collecting it. Many departments, for instance, don’t collect the age of people who are on the receiving end of police officers’ use of force, or they may omit the reason why a suspect was stopped in the first place.

“Within a particular category of data, there are huge disparities in what's filled in,” Goff said.

Katz agreed, “There's no consistent, uniform way of collecting data or agreement on what should be collected.”

There is one part of the Department of Justice tasked with collecting and publishing data: The Bureau of Justice Statistics. But no one argues that the bureau, which is a clearinghouse for all kinds of data on police staffing, prison rape, crime figures and more, should be doing it all by itself. Some, like Katz, believe the answer to improving what we know has to come from individual states.

“I don’t think the BJS can do it,” said John Pfaff, a professor at Fordham Law School in New York. "Every year, Congress asks them to do more and more already. I don't think they have the capacity to do any more. They do amazing stuff, but I don't think they can."

When it comes to bad data, police aren’t even the worst offenders. While there is data on policing and corrections and some on the courts themselves, the biggest piece missing is information on how local prosecutors operate.

"We have really no data whatsoever on what prosecutors do, almost none,” Pfaff said, adding, “We don't know what they're doing, why they're doing it and what drives their decision process."

And that ignorance has an impact on efforts to reduce incarceration levels and lower sentences. Because we don’t have data on how prosecutors work, we don’t focus on them when we talk about reforms, Pfaff said. Gelb called prosecutors “the biggest and most significant black box to be opened in the system.”

The problem with a lack of data on the criminal justice system is more than just budgetary. It’s a cultural issue that gets to the heart of why criminal justice reform is so very difficult.

“For some [police] departments there may be cultural resistance to looking too closely,” Katz said. “Police departments can be very insular, very closed off. Within the closed system they may not even perceive that this may be a best practice.”

This aversion to transparency has rubbed off on lawmakers, who may find the numbers mildly interesting, but not really necessary to guiding policy for a system that largely runs itself, according to Gelb.

“If that's the approach and the attitude, why would you need to have real time, actionable data for policy decisions? Policy makers have not seen the need for it,” he said.

And what we — and policy makers — don’t know about criminal justice could fill a prison.