Early on in What Works in Crime Prevention and Rehabilitation, the new book that informs much of The Trace’s extensive evaluation of policing strategies to reduce gun violence, the volume’s editors — David Weisburd, David Farrington, and Charlotte Gill — cite Robert Martinson’s influential 1974 study for inspiring the long-held belief that no policing strategies work.
Martinson’s focus was prison rehabilitation programs, but his findings casting doubt on the effectiveness of any approach to crime reduction other than punishment rippled through policing and criminology. The upshot, according to the editors: “The idea that crime prevention or crime control could be effective had literally become a radical idea” by the 1990s.
Their book is meant to put that perspective to rest. Weisburd, executive director of George Mason University’s Center for Evidence-Based Crime Policy and one of the nation’s leading criminologists, talked to The Trace about the Martinson study’s legacy, the practical implications of his book’s findings, and what is needed to make the next round of advances in criminology research.
You call your book the first of its kind. How is it different from previous studies on crime prevention strategies?
In the 1970s, going into the 1980s, there was a very strong narrative both among practitioners and among scholars that [prevention] programs didn’t work. In the 1990s, there developed a counterargument. There were a lot of very strong evaluations that began to show that there are things that work.
We said let’s look at all the areas [in crime prevention]. What have we learned? Most of the time there are people who summarize, sort of narratively. We said, “Let’s summarize what we’ve learned from systematic reviews*. What do they tell us?”
And that leads to very, very strong outcomes. It’s almost like the whole thing has come full circle. There is very good evidence that many crime prevention programs in many different areas show strong evidence of success. They can prevent crime.
But that’s not the same as saying everything works.
We’re not arguing that everything is effective, that every program works. But it is really striking how many systematic reviews there are in different areas that show us that crime prevention works.
I think of it this way now: that the Martinson and “nothing works” idea was very naive. Who would believe that nothing works? There must be some things. It would be equally naive to think that everything you do works. And that’s the purpose of evidence-based policy. Because with evidence-based policy, it shows in what direction is the most success.
What should policymakers or police do to put these findings to work?
Look, no one document is going to provide the entire range of solutions that people want. We don’t go into tremendous detail about each specific program, [but the book lists the underlying research, so a reader can] very quickly get a sense of where are the areas where things seem to be useful.
In policing, for example, focusing in on micro-geographic places is a very good direction. And there are lots of studies that reinforce that now using different approaches.
One approach that your book says is effective is targeted patrols to reduce illegal possession and carrying of firearms. Is that another name for stop-and-frisk?
It’s not directly on stop-question-frisk. But there have been a series of approaches that would focus, for example, on hot spots, and stopping people to check for guns. And those are more highly focused than most of the stop-question-frisk programs. The interventions that show strong positive effects are very, very focused. Let’s say you have a street with a large number of shootings over the last month or two. Would it be reasonable on that street to stop everyone — if legally allowable, of course — in that situation?
So in other words, stop-question-frisk as a general strategy in a city might be a very bad idea. It can lead to all sorts of negative unintended consequences. I think the New York strategy was too broad, and too general, to really be justifiable [in light of its racially biased outcomes]. But maybe in a situation where you highly focus on the specific places where there are very serious problems, it may be a useful strategy. It may be justified.
What other policing strategies have a major impact on gun violence specifically?
Another area is focused deterrence, where you say, “I’m going to focus in on certain places and certain high-rate offenders, and I’m going to use a lot of levers. I’m going to do a lot of things to try to ameliorate these problems.” In the systematic reviews of focused deterrence, which Anthony Braga and I did, we found very, very strong effects with focused deterrence. We warn, however, that the studies there tend to be a little weaker in design. There aren’t any randomized experiments. With hot spots policing, there’s also a positive finding, but the declines in crime are a bit more modest in terms of the size of the effect. But either way, you’re getting significant outcomes.
You note in your book that a lot of gaps remain in the research on policing strategies. What needs to happen for us to learn more about what works?
This [research] has been a good investment. Now let’s invest more in criminal justice research.
For medical research, the National Institutes of Health has a budget of about $30 billion a year. Not all of that is for outcome studies or things of this sort. But they have very, very deep pockets. The total amount of research funds in the federal government for criminal justice I suspect is between a hundred and a hundred and fifty million dollars. Now obviously you’ll achieve a hell of a lot more if you invest more in it. So, yeah, we’ve learned an awful lot. But we really need to take another grand step forward to get where we have to go. For that to happen, there’s going to have to be a much larger commitment from the federal government.
*In a systematic review, researchers methodically survey dozens or hundreds of studies on a given topic to draw conclusions from a large data set.