LPRC - CrimeScience Episode 66 – Science & Evidence-Based Policing featuring Dr. David Weisburd (George Mason University) Part 2

Episode Date: April 1, 2021

Dr. David Weisburd, Distinguished Professor at George Mason University and Executive Director of the Center for Evidence-Based Crime Policy, joins Dr. Read Hayes on LPRC CrimeScience to discuss science in evidence-based policing, the value and challenges of real-world research, place and crime event clustering, and much more. This is the second and final part of the discussion. The post CrimeScience Episode 66 – Science & Evidence-Based Policing featuring Dr. David Weisburd (George Mason University) Part 2 appeared first on Loss Prevention Research Council.

Hi everyone, welcome to Crime Science. In this podcast, we aim to explore the science of crime and the practical application of this science for loss prevention and asset protection practitioners, as well as other professionals. We would like to thank Bosch for making this episode possible. Be a leader in loss prevention by implementing integrated solutions that enhance safety, reduce shrink, and help to improve merchandising, operations, and customer service. operations, and customer service. Bosch integrated security and communication solutions span zones one through four in the LPRC zones of influence while enriching the customer experience and delivering valuable data to help increase retail profitability. Learn more by visiting Bosch online at boschsecurity.com. This is the second part of our discussion with David Weisberg and our host, Dr. Reid Hayes. I think one of the things too, I wanted to ask you about, you know, all, so much of what you are learning and putting out to the rest of us around crime

events and how they're clustering and why, and the challenges there, but the opportunities because of understanding place and crime. And I've watched as you've gone to smaller and smaller and smaller environments, you know, macro, meso, micro, and even below that, if you will. Can you talk a little bit about crime event clustering causes, challenges, opportunities, David, at all? Look, you know, I first saw this when I took a job right before I finished my PhD at the Bureau of Justice. And I was evaluating community policing program. And each community police officer, there were nine of them, they each had a beat. It was 20 to 30 or 40 square blocks.

And those were considered bad parts of town. This was in New York in the 72nd precinct in the 1980s. And it was bad in a way that fortunately things have in many of these places improved a great deal. It was really, there was a lot of drugs, prostitution, violence, lots of disorder, et cetera. And I would walk with the cops in these supposedly bad parts of town, and what I began to see was we spent all our time on one or two blocks. And that, for me, is the origins of the idea of hotspots. In other words, in a so-called bad neighborhood, only a few streets generate most of the problems. Now, across my career, I then realized, you know, I began to look at that. One of the early

projects was with Larry Sherman, where we looked to see whether the police, if they focus on these one street hotspots, could have an impact. And we did in the Minneapolis hotspots experiment. observational, or statistical research. And another, if you like, that experience when I was much younger in the 72nd Precinct was a eureka moment. In other words, that most crime, you know, what I learned in school was that there were bad neighbors of town. And I learned walking the street with cops was that there were actually a few bad streets in that part of town. A few years ago, about 10 years ago, so I began to put together research I was doing across different places. And I'd done a study in Seattle. And what I found was in Seattle, that 5% of the streets produce 50% of the crime for 16 years. And that's kind of amazing. Every year, pretty much the same concentration level, 5% of the streets, 50% of the crime. And that was during a period when there was a drop in overall crime of about 23% in Seattle. So even during the crime drop, every year crime might have been dropping, but about the same proportion of places, a relatively small proportion of places, produce 50% of the crime.

Now, right after I'd done that study, I did a study in Tel Aviv, and using the same methodology, which is looking at street segments from intersection to intersection. And in Tel Aviv, 5% of the street segments produced 50% of the crime in the year we were looking at. And 1% produced 25% of the crime, which was pretty much the same figure in Seattle. And for me, that was a eureka moment. It said to me, and I developed the idea from that of the law of crime concentration. In other words, it's not only the crime is relatively concentrated on a small number of streets in the city, it seems to be concentrated, at least in larger cities, at about the same level of

concentration. And that's a really important idea, I think, in terms of reorienting the way we think. Now, I should note that when I began my career, most people thought that the only way you could do something about crime was pretty much to focus on individuals and why people commit crime and doing something about them. And there was some work on communities in crime, but individuals really were the focus. And this work on places led me to the idea that one way we can capitalize to do something about crime is by focusing on microgeographies, on these single streets or a small group of streets

that produce a large proportion of the crime problem. You can get greater effectiveness and greater efficiency by doing that. that produce a large proportion of the crime problem. You can get greater effectiveness and greater efficiency by doing that. And that led to a whole series of experiments, as you know, in which we randomly allocated policing to hotspots. Up until the time when Larry Sherman ran our first experiment, the assumption was the police couldn't do anything about crime. And we didn't think that could be right. And we decided, both of us from our previous research experiences, mine in New York and his in Minneapolis, that if we could focus police on these places that have a lot of crime, these hotspots, that we would see a determined value. And that's been shown through more than 70 strongly designed experimental and quasi-experimental studies. So this focusing in is very useful.

And I suspect that it applies to many, many other sorts of problems. I suspect something like 5% of the stores produce 50% of the theft in stores. It'd be good to look at. There's also an idea that's come from a business in economics called the Pareto principle, which identifies, I believe, that 20% of factories produce 80% of production. I think this sort of idea of concentration is widely spread across areas and suggest the importance of focusing in. Usually problems are very, very concentrated, and that concentration provides an opportunity to be efficient, effective, and focused in how you respond to those problems.

And that's just an excellent description, and it really sets it up. It is a big part of what we try and work with our constituency on. And you know, it's really interesting, David, because yes, absolutely crime, theft. Let's take theft, shoplifting, for example. Of course, it does cluster a very small percent account for the majority or at least plurality of the problem. But then we, you know, we're looking at these micro environments and you see a couple other clusters I was going to just bring up.

One is, you know, well, they're targeted item clusters. And so, you know, people are going to, let's take a Walmart or Target, or example, blades and razors, particularly Gillette, they're highly popular,

highly stolen. And so you're going to see a cluster of items targeted or the targets are clustered in those areas. Let's say health and beauty care, electronics. And then you can continue to drill down, drill down, even the specific SKUs. It might be eight count or four count, this type within this brand and so on. But the other is, uh, is the theft activity clusters. So you might take your blade pack from this area or your whatever, uh, premium shave item. And then, but you're going to probably conceal it in another area. Maybe that's because

it's more low observability, for example. So we, even if you used old fashioned push pin maps, you see, okay, here's all the cluster of what we're losing and where they're concealing it or where they're pushing carts out. You might have three entry exits at a Walmart supercenter, but they're going through the garden center. I wonder why most are going through there. Well, it's low observability. So the principles that you bring up when you've worked so much are so helpful. One way to think about what you're talking about is that within any particular environment, there usually is a clustered area. So someone did in Sweden a study of malls, and they found, not surprisingly, that within a mall, it's a very small number of shops that produce most of the problems. Within a mall, it's a very small number of shops that produce most of the problems.

I think that you could say that each one of these environments, a store, for example, a store has many units of environment, if you like. And there are probably a relatively small number of those units that are producing a good deal of the crime problem. I think what's put people off from taking this sort of thought in the past was the idea they said, well, if I just crack down on that street or that place or that product, people are just going to steal other things or going to commit the problems or create problems in other places. Some of our work from a study I did in Jersey City where we examined this pretty carefully was that when you focus down on a particular environment that has problems, quite often crime doesn't just move around the corner. In fact, the areas nearby tend to get better because around the corner does not necessarily have all the same opportunities. In a store, for example, I suspect that one area where they take from, there might not be many eyes

on it, right? People might not be watching it. It can't move to a place where there's eyes on it because they'll get caught. In real life places, there's also the element that the offenders have familiarity. Offenders get familiarity with certain places or contexts. They don't like to move very much because that involves risk. When you move, you might hit an environment you don't know. It reminds me when I'm driving places. My wife is always saying, let's try someplace else to park, whatever. And I'm always saying, let's go to the same place. I know it. I'm unlikely to go down the wrong way on a runway street and these sorts of things.

So familiarity is very important. And that's another way of saying that you have to think about the way people offend the same way we do all our different activities in life. There's area of comfort, what makes us comfortable, our fear of risk, efficiency and effectiveness, what's going to be the best place to do it, etc. Now, I like it. In our case, too, places that they can convert stolen goods to cash, the proximity and, yeah, their knowledge, they feel comfortable that they won't be victimized. And all those things seem to come into play. This is, again, you know, people have a tendency to

think of offenders, especially offenders who seem to them to be very driven, like drug users or prostitutes or other sorts of activities like that. We found in our Jersey City study that the police had cracked down on a very high activity prostitution site and there were other prostitution sites not that far away. We did some qualitative work and wanted to understand why they didn't just move to another spot, right? And one of the prostitutes said, he said, well, I don't feel comfortable with those girls over there. And I think we have to remember that people that commit crimes are just like everybody else.

They're just people. He doesn't feel comfortable with those girls there. It's not a comfortable place for her. Among drug dealers, there was a different sense. When the drug dealers were asked why they might not have moved someplace else, they would say things like, oh, there are people there, man. If I go there, I'll get shot. Or they'll say, an interesting thing that always struck me was they'd say, you know what? I don't know the people there. I don't know who's good and I don't know who's bad. And they didn't mean

what you and I mean by good and bad. They meant they didn't know who was going to call the police on them or not. Right. They didn't know. And they didn't want to make a mistake. This is interesting because among policing, there was a movement in the 1970s to avoid corruption of having police officers shift from beat to beat, not get too familiar. So they couldn't make, you know, strong relationships with people. So they couldn't make, you know, strong relationships with people. And one of the results of that was the cops were making all kinds of mistakes. You know, they didn't know the people. So that a guy would be walking down the street, he acts a little weird, and they'd pick him up. Turned out this is just a guy who has some issues or whatever, you know. So, you know, in a way, when they created those rules

to try to reduce corruption, they increased other sorts of problems because familiarity is an important part to work of all sorts, illegal and legal. That's amazing. So maybe this is good. So if we could maybe go over another area that I really think is powerful that you're working on, and that is, especially new right now, talking about policing, okay, because this is so transferable to what the other area that I'm mainly involved in, positively engaging with community members, community policing, and all the ideas over the years, problem-oriented policing, and things that you very well know and have helped improve. That may be a different construct, if I'm reading some of your work, than what you do to shape

environments, help victims become less vulnerable, but suppressing high-impact offender behavior. In other words, police, law enforcement versus community bonding, that they may be somewhat different, not mutually exclusive necessarily. Can you talk a little bit about that, David, and what that might mean for us? Part of my thinking, this has been influenced by a National Academy of Sciences Committee on Proactive Policing that I chaired a few years ago. And we reviewed the evidence about different proactive policing strategies, and we included strategies like community policing or procedural justice policing, in other words, that were seen by many of their proponents as a method of reducing crime, but also as a method of improving relationships with the public. There are, on the positive side, a bunch of proactive policing strategies that were found to be effective through good science, hotspots policing, third-party policing, focused deterrence, and others as well.

Those tactics tended not to have any effect on community attitudes, on the people who live in those streets or even the larger community. On the opposite side, community policing and procedural justice policing and broken windows policing, which all seek to sort of make people stronger, to strengthen the community, to increase positive relations between the police and the public, at least procedural justice policing, increased positive relations between the police and the public, at least procedural justice policing, community policing, they don't have much effect on crime that we could see. So it's almost like you have two separate trends here. You have a group of strategies that seem to reduce crime, and a group of strategies, community policing in particular, that improve relations with the public but don't seem to have much effect on crime, right? So they're like on two parallel universes. And the only place they cross is in problem-oriented policing. And

the reason for that, we think, is because many of the problem-oriented policing strategies integrate community or community engagement and involvement as part of what they're doing. And that led me to an idea. I think it's not a radical idea, but it reorients the way the police think about these sorts of problems. The evidence is, for the most part, that effective crime control does not necessarily increase citizen evaluations of police legitimacy. Hotspots policing, other sorts of activities, they didn't increase positive attitudes of the public towards the police. I should note they didn't decrease those attitudes. Many people see them as creating problems. They don't. There isn't good evidence outside of stop,

question, and frisk that these proactive policing strategies have negative effects on the public, but they don't have positive effects either. And that's a very important lesson for the police to, to police executives to learn because police executives think if I just do a good job in reducing crime, the public is going to evaluate us positively. And that doesn't seem to be true. That doesn't mean the police should not do a good job and do something about crime. The job of the police,

one of the main jobs is to doing something about crime. The job of the police, one of the main jobs, is to do something about crime. And therefore it's important they do, and that affects their budgets and how the mayor sees them, et cetera. But it's important to recognize that that is not the path, at least in the research we've seen, to gain positive evaluations of the police, increasing evaluations of police legitimacy. At the same time, community policing does seem to lead to those outcomes of positive community aspects. So in my view, the police should be tackling these problems as two key problems for the police. The police, their job is, one of their key jobs is to reduce crime. It's a democratic society. They're supposed to do that with the support and legitimacy of the

public, and hopefully that will help them in the long run. But to achieve those two things, the police may have to use different types of programs or integrate different types of programs, which is where the problem with the policing idea, I think, adds knowledge. Because I think when police do hotspots policing, they want to integrate community policing, perhaps also procedural justice policing into the activities. But they have to recognize that it's not just about controlling crime. They also want to support the public. We're democratic police agencies. That's an important value they have to encourage. And that means they may have to use different strategies and combine them in different ways. The real important thing is do not assume that just because you reduce crime, that will

raise the evaluations of the public or the police. That's great discussion. And again, I see the parallels on the private side that we deal with that the asset protection or loss prevention leader, your real role is to protect the vulnerable, of course, people in places, but to safeguard, but overwhelmingly to support the success of the enterprise, total enterprise success. And so I think the same thing in the city. And I'll never forget, it was when Bratton was still, it was at the end, but his last term as the commissioner in NYPD, we were up there at an annual conference that takes place every January, all these retail APLP, as we call them up there, he spoke, but he had this table in front of me. He said, look, this is New York City. This is an ecosystem and all this is going on. My role in my department's role is to support healthy function of this ecosystem, not to put people in jail, not to enforce laws, but to support the healthy.

You know, I'm paraphrasing, of course, but it seems to go back to what you were just talking about, Dave. back to what you were just talking about, Dave. I think Bratton, especially as time went on, began to understand the importance of not just doing something about crime, but also making sure that the underlying attitudes towards the police were positive. But there was some criticism in Comstock, perhaps after Bratton, that they became so focused on the crime issues that they neglected their relationships with the public. And you can see that a bit in some of the reactions these days to the police in New York City. Even before the recent events, Floyd and others that have raised all sorts of concerns in

some people's minds about the police. I was in New York City at a meeting of their criminal justice division, and it was open to the public. And I saw the kinds of gut feelings people had against the police, like defunding the police was getting a lot of applause at the time. I don't think that's a realistic, good thing for communities. But it just points to the fact that the police have to consider not only the issue of doing something about crime and preventing those sorts of problems, but also making sure they continue to engage and get the support of the public. I mean, that's democratic policing. That's who pays the salaries.

Is there a place in the United States, David, for the idea that the UK has, and I'm probably going to get this wrong, the policing college, but my view a little bit of the NIJ is the funding agency in part for academics to conduct the research that they want to conduct and that may or may not be useful or helpful or it's basic science or whatever it might be. But a lot of times it does, it seems like some of the agencies I work with, they're initially not hostile, but they're on guard. Okay, what is this person trying to find out or do or what agenda might they have other than to help us protect our people

or safeguard our city? Any thoughts on that? Is there something that makes sense, not necessarily another bureaucracy, but an agency that might help law enforcement or policing get better and want to get better? Look, I think there are a variety of problems the way things are organized now. We can see a little bit, we can see indications of that in terms of the reactions of the public and the situation we've had over the last year. Policing needs an agency devoted to policing, irrespective of what NIJ is doing now.

Policing needs a police college or a funding agency and practice, a place where they would both do research about the police and for the police, and that would also develop rules and ideas and standards for police departments across the country. We right now have 16,000 police departments that are essentially all independent. That's not good. We need more control, if you like, or at least more standards set by the federal government, as they do in a place like the UK and with the police in college. Yeah, I think sometimes the NIJ, when there's been reports about NIJ, they've looked to the National Science Foundation for models of how to do research. And I

suspect they probably should have been looking at NIH for models of how to do research. At NIH, there are enormous amounts of money spent on how to. How do you treat patients? What's the best and most effective way to do it? There's also enormous amounts of money spent on learning about different dosages, levels of treatment, etc. The National Institutes of Health have a budget of, I believe, over $40 billion a year. And the National Institute of Justice has a budget for all areas, including policing, of about $100 million when you throw everything in, I believe, a year. So there's a real problem. But we need a, I think you're exactly right, Reid, the last year convinces me we need a national organization, a policing college, that would develop research, practical research for policing, basic research about policing, that would create standards for policing in the field, that would become a real backbone for building a better policing in the United States over the next 10 years or 20 years, which is really needed. But that will take a large investment. That's very hard to get

from politicians. It will take a much larger investment than they have now. Not a hundred million, maybe not 40 billion, maybe 10 billion, but a large investment in the creation of a agency in which the police are the central component. I think that's very important for policing in the future. Excellent. So David, what have I not asked you? What might police practitioners, loss prevention David, what have I not asked you? What might police practitioners, loss prevention practitioners, what might budding or seasoned academics need to know while we're talking? Look, I'll leave you with one thing that I've been doing lately, one area. We know that it's a well-established fact at this point that crime is concentrated at a relatively small

number of places in the city, which we call hotspots. And the primary ways we've tried to deal with those hotspots has been through policing. I should note to some degree work like yours have tried to deal with the idea of hotspots through opportunity reduction mechanisms and issues of that sort. But I did a very large study for the National Institutes of Health on hotspots. And what I found, because I wanted to understand not just issues of deterrence and the traditional data we looked at, but I wanted direct measures of what's going on with the people in those places. And one of the most interesting findings, and one that's being published in a paper now coming out in Prevention Science,

and one that's being published in a paper now coming out in Prevention Science, is that if you want to understand changes in crime and hotspots versus non-hotspots, if you like, over a period of time, many of the things we've been looking at, issues of opportunities, are very, very important. But it also turned out that what Rob Sampson has called collective efficacy is also very important. If people on a street are more willing to, are more integrated with each other, trust each other more, and accordingly more willing to be involved with doing something about their problems on the street, then that reduces crime. This is an observational study, not an experiment yet, the next step of experiments. But that suggests that there may be other approaches we could use besides law enforcement that would also do something about crime.

And perhaps focusing a bit on informal social controls, as sociologists I think would call it, may provide opportunities. So I've become very interested lately in how can we increase collective efficacy on streets, how we have people to trust their neighbors more, how can we get them to be more interested in doing something about problems. And I think if we could increase collective efficacy on these problem streets, we would not only get benefits in terms of informal social controls, but they would also do a better job in working with the police in terms of doing something about crime. And then I guess that leads to a more general idea that I think it's important to do some thinking about what else besides policing could we use to improve places? What other opportunities do we have for reducing crime? Policing is very expensive.

There are some elements of policing that also can have negative outcomes, arrests and stops and issues of this sort. So are there other ways we can add to police deterrence and police activities? Are there other ways that we can do something about the crime problem and take advantage of the efficiency of dealing with crime hotspots. Powerful, powerful message. And ironically, by the way, last January, right before the pandemic hit in earnest, we had our advisory panel here in Gainesville on campus.

And we went through with them some of the leading ideas. And they had one was an erosion of consequence for offenders that they had sincere concerns about and would like to see research at all levels. But another one was they don't know and use the term collective efficacy, even though we've broached it with them. But a collective action, in this case by local business or even if they're chain stores, that they could work collectively for a few reasons. One, they really can't afford to take all the opportunity reduction measures that they would like to, but collectively they might, and they might get better. And so how can they work collectively with each other? And this is their idea, in a strip center, in a mall, or in a in a smaller environmental area, and work with law enforcement or policing as well, but get to do what you're talking about a little bit, David, engage with the locals. How can we have reading programs?

How do we do this? How do we engage locally? But, but we want to do, they want to do something that really has merit. And to me, the collective efficacy argument and Samson's work and where it sounds like you're really heading was a big idea that they latched onto, not as articulated as well as what you're doing right now. So I'm looking forward to what you put out on this. Yeah. Sometimes the problem, I guess, with some sociological concepts is it gets a little jargony, right? Collective FC, you're talking about collective FC, these guys, if these guys who have stores would get together, trust each other, and

work together to do something about the problems, they're going to be a lot more effective. And that's informal social controls, right? And that will work better with the police and other agencies. So I think it's an idea that has merit. I don't believe in defunding the police, but it's an idea that has merit. We should look for other opportunities as well about doing something about crime. I like it. And I'm not even sure about reforming the police. I think it's the ideas around what you've talked about today. And that is, you know, better science, you know, focused policing. But the science shouldn't be scary.

The science goes back to logic models, starts with collective efficacy, and maybe works up and down at those strata. So that's an exciting prospect and something that hopefully is common sense and achieves the outcomes we want. very informative. It's just one more step in the intentional or unintentional mentoring that you provide, at least for me and for my team here as well, David. And I can't thank you enough for your time today and all your insights. A pleasure speaking to you, Reed. I enjoyed it. And I'll tell you what, in 2020, you've heard it a hundred thousand times maybe, but, you know, stay safe, best wishes and happy holidays to you and yours. And so thank you everybody for tuning in to another episode of Crime Science, the podcast. Thanks for listening to the Crime Science podcast presented by the Laws Prevention Research Council and sponsored by Bosch Security. If you enjoyed today's episode, you can find more Crime Science episodes and valuable information at lpresearch.org the content provided in the crime science podcast is for informational purposes

only and is not a substitute for legal financial or other advice views expressed by guests of the crime science podcast are those of the authors and do not reflect the opinions or positions of the last prevention research council