Social Network Analysis Beat Intro Voiceover 00:00 This is the Beat—a podcast series that keeps you in the know about the latest community policing topics facing our nation. Interview Barry Bratburd 00:08 This is Barry Bratburd with the COPS Office. With us today is Dr. Andrew Papachristos, professor at Yale University. Dr. Papachristos is here today to talk about social network analysis. Now, some people hear “social network analysis” and mistakenly think it has something to do with Facebook or other social media sites. What is social network analysis and how has it been used in policing as well as other fields? Dr. Andrew Papachristos 00:32 So I think the linking social network analysis with Facebook and Twitter and stuff is the popular, knee-jerk reaction, but it’s not actually wrong. The science behind it, who suggests who your friends are, which books you should read, is actually based on a larger science that is social network analysis. Social network analysis has been around for about 70 years or so; sociologists have been doing it for that long, and anthropologists. Over the last 20 years, you’ve seen physicists and computer scientists and other people really take off with it. As a field of science, what it does is it analyzes the relationships among people or groups or organizations and how those linkages—how those relationships among groups—affects all sorts of behaviors. You’re interested in studying the structure of the network itself—is it a very dense cluster of people? Is it a diffused cluster? And how those shapes of the system affects behaviors like voting, who gets shot, who gets a disease, how you perform on your evaluation, and so on. In terms of the criminal justice field—it’s actually pretty new to the criminal justice field. It’s been used pretty heavily in public health since the AIDS epidemic and understanding how diseases are transmitted and how you can use it for prevention purposes. It’s been used in business for the last 20 years to understand evaluations and how ideas and collaborations happen. It’s been popularized by Malcolm Gladwell’s book, The Tipping Point, how things spread. It’s new in the criminal justice field, very new, and in fact the criminal justice field really lags behind in this network revolution, in part because of how we think about and do our jobs but also in part in terms of how technology makes its way into policing and criminal justice. Barry 02:14 You make a good point. It seems that social network analysis has yet to be widely adopted by police practitioners. How do you think it could become more widely used by crime analysts in the future? Andrew 02:24 So I think moving forward, the path social network analysis has to take is really similar to what geographic hot spot mapping did in the 80s and 90s, which is first, understand what it is and how it can be helpful. In most instances, it’s taking things that every beat cop, street worker, minister worth their salt already knows and aggregating up. In some ways, to continue the public health analogy, when you look at an STD outbreak, you want to know who’s sleeping with whom. Who are the sex workers and long-distance truck drivers who spread these issues. If you ask each one of those persons—each one of those people—who they slept with, they would tell you two people, which is always what they tell you, then you ask those two people who they slept with and they would each tell you two people. But the person who’s working with that one sex worker only focuses on one person. Same thing with a good beat cop. They’re working a particular case, an individual. Any good outreach worker is working their caseload. They don’t often see how these linkages affect the people they’re working with, because that’s just not how, but it still affects your life. Just like your partner’s ex sex partner can give you an STD indirectly, same thing with offenders and people in the criminal justice system. I think first it takes understanding that logic. Second, it takes developing the tools to really get it in the hands of the people that are going to use it. But it has to be something that they need and want to use, it can’t just be something that was forced on them. It has to be developed with them in mind. Barry 03:56 Now recently there’s been a focus on understanding violence epidemiologically, that is, understanding how violent behavior is transmitted through populations. How can social network analysis help us get a better picture of how that happens? Andrew 04:09 Not to keep talking about sex, but the analogy to sex is really the right one, which is, if violence is a disease, how does it spread, is the real question. Most models use what we call “risk factors.” Basically, the dumbed-down version of the fancy statistical equation is if you’re poor, young, black, live in this neighborhood, don’t have a job or a high school diploma, you’re going to be a victim or an offender. The problem is most poor young black people that live in those neighborhoods never shoot anybody or are never shot. The former analogy of just looking at risk factors assumes that crime is airborne—you walk into a neighborhood and you get shot or you walk into a neighborhood and you start shooting people—but it’s more like a sexually transmitted disease where it’s more concentrated. It’s who you mess around with and who you hang with that really puts you at risk and not just having these characteristics. That’s what I see as the big contribution that makes this parallel between criminal justice and public health, is how do we understand how the disease, crime, diffuses? It takes a particular shape and form, just like any other epidemic. It might not be an airborne pathogen but rather a blood-borne pathogen. Those are very different ways. Is it the flu or is it hepatitis? How do we then design interventions around that, is the next step. Barry 05:30 Well, finally, we know that you’re now with Yale University. What are you going to be doing at Yale? What are you going to be focusing on? Andrew 05:37 I’m going to be focusing my time at Yale on really expanding this network thinking in a couple of ways. The first is in scope; I’m going from just working on Chicago to working with about five or six different cities right now in the U.S., including Cincinnati, Oakland, Boston, Newark, and a few others—that’s the first step—and really understand these epidemics in their context. Each one of those cities is different. Each one of those cities has different problems in relationship to guns and gangs. I think what the network approach lets you do is see what are the commonalities and what are the differences and pinpoint them in terms of prevention efforts. The second thing is really to expand this line of thinking, both from developing a set of tools for enforcement and prevention efforts, but also getting that out there in how people do their jobs. Hopefully, again, all of my work which I haven’t said, has been tied to prevention efforts or enforcement efforts, and making sure that we build better efforts. Building better prevention efforts so that we can not just simply do less harm but do more good. Barry 06:45 Great! Well thank you so much for being with us today. We really appreciate you providing us with your expertise and your time. Andrew 06:51 Great. Beat Exit Voiceover: 06:52 The Beat was brought to you by the United States Department of Justice COPS Office. The COPS Office helps to keep our nation’s communities safe by giving grants to law enforcement agencies, developing community policing publications, developing partnerships, and solving problems. Disclaimer: 07:09 The opinions contained herein are those of the authors and do not necessarily represent the official position or polices of the U.S. Department of Justice. References to specific agencies, companies, products, or services should not be considered an endorsement by the authors or the U.S. Department of Justice. Rather, the references are illustrations to supplement discussion of the issues. ####END OF TRANSCRIPT####