Does Evidence Matter in Justice Policymaking?

Print More

Photo by John Lord via Flickr

For two decades, criminal justice advocates have been talking up the idea of basing anticrime policy on scientific evidence.

How much is it actually happening nationwide?

That question was on the table Thursday at the American Society of Criminology’s annual meeting, held this year in Philadelphia. Criminologists long have complained that policymakers tend to ignore their studies and pursue ideas based more on whims than science.

Laurie Robinson, former Assistant U.S. Attorney General now on the faculty of George Mason University, believes there has been much progress but also a lot of resistance to the idea of backing up justice policy with solid research.

In the first of two Robinson stints at DOJ, a report assessing what works in fighting crime and what doesn’t helped her cut federal funding for programs like Drug Abuse Resistance Education (DARE) and military-style boot camps for low-level offenders.

Still, dubious ideas like gun buy-backs by police agencies keep recurring even though studies have found them ineffective.

“Science has a hard time combatting emotionally popular programs,” Robinson said during a panel discussing the topic.

Edward Mulvey of the University of Pittsburgh, who heads a Science Advisory Board at DOJ, agreed that evidence on the spread of evidence-based programs is mixed.

Many “unsound policies” remain in the criminal justice world, partly because much of the public doesn’t see the value of waiting for evidence to justify a policy change, Mulvey said.

He takes the “long view” that proved practices eventually will prevail over “media headlines” about ideas that prove ineffective.

The Trump administration has said that it will retain the science board at OJP, which was established by former Attorney General Eric Holder.

At the National Institute of Justice (NIJ), DOJ’s research agency, David Muhlhausen has moved from the Heritage Foundation to become director.

At Thursday’s criminology program, Muhlhausen declared that “science-based crime policy is on the rise, but we need to improve — we’re not where we want to be.”

Mulhausen is enthusiastic about a website established by Robinson,, which assesses the effectiveness of many anticrime programs that have been studied.

Muhlhausen’s primary concern is that there are too many program evaluations that are “quasi experimental” — far from definitive because they weren’t done using the “randomized controlled trials” in which people getting an experimental treatment are compared with similar groups who aren’t subjected to it.

He cited the example of drug courts, which he said had repeatedly been evaluated using the “quasi experimental” method.

The new NIJ director said that in general, he wanted to stop funding government-subsidized programs that don’t work, to avoid a “waste of taxpayers’ money.”

His criticism wasn’t limited to the Justice Department.

Muhlhausen cited a project of the Department of Labor supporting job training for former prisoners that the agency touted while not disclosing that a randomized controlled trial showed it was ineffective.

He also cited the Hawaii-based HOPE program (Hawaii Opportunity Probation with Enforcement), which puts some criminal defendants on probation under the threat of quick punishment if they violate rules.

Muhlhausen said an initial evaluation in Hawaii found the program valuable but randomized controlled trials in other states cast doubt on it.

“We have to be careful to define ‘what works,'” Muhlhausen said Thursday.

Muhlhausen admitted that evidence-based anticrime policies would be a “tough sell” to some audiences, such as working police officers.

He is supporting a project to instill academic concepts more widely among the ranks of criminal justice practitioners, a group he dubbed “pracademics.”

One leading justice practitioner who agreed that it can be difficult to instill evidentiary principles in the work of police and other criminal justice workers was Gil Kerlikowske, a former Seattle police chief and director of National Drug Policy under President Obama.

Kerlikowske noted that many large police departments had improved their techniques in such areas as videotaping confessions and obtaining witness identifications of crime suspects, but that many smaller departments had not caught up with needed changes. He said the academic community bears some of the blame for not offering their expertise to small police agencies.

Criminologists seemed pleased that NIJ’s Muhlhausen had embraced evidence-based policymaking in a presidential administration that has shunned scientific evidence in areas such as climate change.

Still, Alfred Blumstein of Carnegie Mellon University expressed doubt that Muhlhausen could insist on the “gold standard” of randomized controlled trials for most studies of anticrime projects.

Important areas such as the death penalty aren’t appropriate for such experiments, Blumstein said.

Muhlhausen agreed that every crime study couldn’t be a randomized controlled trial. He repeatedly said that he wants NIJ to “advance the ball” and not to fund repeated studies that don’t aim to break significant new ground.

Ted Gest is president of Criminal Justice Journalists and Washington Bureau Chief of The Crime Report.

Leave a Reply

Your email address will not be published. Required fields are marked *