Solutions in Corrections: Using Evidence-based Knowledge

Kristina Rose: Good morning and welcome to today's session in NIJ's Research for the Real World seminar series. My name is Kristina Rose, I'm the Acting Director of NIJ and I'm very happy to see such a good crowd here this morning. Today's lecture is titled, “Solutions in Corrections: Using Evidence-Based Knowledge,” and it features Professor Ed Latessa. He is the director of the School of Criminal Justice at the University of Cincinnati.

And in order to get us started today, I'd like to introduce our Assistant Attorney General, Laurie Robinson. She was sworn in in November of 2009 and we're very grateful to have her here at OJP. Under Laurie's leadership we have renewed our commitment to evidence-based programs and practice and policies that are both practical and useful in the field. This seminar series is an integral part of that commitment. We're fortunate that Laurie has taken the time out of her busy schedule to be with us today, so please join me in welcoming Assistant Attorney General of OJP Laurie Robinson.

[Applause.]

Laurie Robinson: Good morning, and it's great, as Kris said, to see a good audience out here this morning. So I want to join Kris as well in welcoming all of you to this installment of NIJ's Research for the Real World series. And I also want to thank Kris, Thom, Ellen and Yolanda, and actually all of the NIJ staff who've worked in putting these sessions together. It's really a great series as you'll see from the presentation today. And I'm just delighted that we have Dr. Ed Latessa from the School of Criminal Justice at the University of Cincinnati here with us this morning.

 

 

And if I could just quote from one of his writings. He published an article six years ago called, “The Challenge of Change: Correctional Programs and Evidence-Based Practices.” And in it he said, “If we want to have an impact, it's incumbent on us to translate research and findings into understandable concepts and terms, and then to present them in a way that helps practitioners understand the value of the research to what they do.” And Ed, I couldn't agree more with what you said there. While we're working so hard to build up our base of knowledge as NIJ is doing, and as OJP is doing through our evidence integration initiative, I think we're all well served to heed your advice. We should be thinking as much about building and bridging the gap between research and practice as we are about the research itself. And in fact that's really the point of E2I [evidence integration initiative].

And let me just say that I think you were actually preaching to the choir. NIJ has certainly been a real leader in getting knowledge out to the field. But we still need more voices like Ed Latessa's out there because we have a great deal more work to be done here to make sure that research is being heeded and is being applied. So I'm very glad that Ed Latessa could join us today to share some of the lessons he's learned. So please join me in welcoming our guest, Dr. Ed Latessa.

[Applause.]

Edward Latessa: OK. Thank you Laurie. I want to thank Laurie and Kris and Thom in particular for organizing this event and for asking me here. I'm not a big podium guy so I'm not going to stand behind the podium, if you don't mind; I'd prefer to walk around. I find it's easier to intimidate students that way.

[Laughter.]

Latessa: I've only got about an hour here to talk, and them I'm going to take questions and we can have some interaction so I'm going to move through the material rather quickly.

But as I thought about the presentation, and I've done these quite a few times around the country, and I could go through all the data and all the studies. But rather than do that, I thought I'd do it a little bit different, we could have a little bit of fun. And I entitled it, kind of, the lessons I've learned over the years evaluating correctional programs. I've been doing this work for over 30 years now. I thought about it the other day, actually, about 35 years. I was a graduate student at Ohio State and I worked for a man named Harry Allen, and Harry ran a research shop at Ohio State that was kind of still the LEAA days, and there was a lot of money coming out for correctional programs and one of the requirements was you had to be evaluated.

You had to build in a percentage and folks would call us and we'd take the contract and we'd evaluate a correctional program. They were typical outcome studies and pretty fairly mundane but they went something like this: an agency would call and they'd say, “We have a grant from so and so and we have to do evaluation and we want you to come and do it for us.” And so they'd send me down to meet with the agency folks and I'd sit down with them and I'd say, “Tell me what you're going to do.” And they'd say, “Well, we're going to carefully screen these offenders and we're going to assess them very well and we're going to put them in this great program.” And I'd write all that down like I cared about it. And they'd say, “And we're going to send them out and they're going to do, you know, be rehabilitated,” and I'd write that down.

And then of course I was just concerned about getting the comparison group. Where are we going to find it? Where are we going to get the data? And so forth. And we'd do the study and sometimes the program would have some effect and everybody was happy. And more often than not it didn't work, in which case it was obviously my fault as the researcher. The problem was, if the program worked, we weren't sure why. And if the program didn't work, we weren't sure why. And then of course we started doing more process evaluations, looking at what kind of, you know, did they do what they say they were going to do? Did they do it on time? Unfortunately, oftentimes they did, and it still didn't make any difference. And so over the years my work has evolved. That's what I want to talk a little bit about, some of the lessons I've learned here over the years in evaluating correctional programs.

The first lesson is that some things don't work. This really came to me probably about 20 some years ago when I was asked to evaluate a program in a county not far from Cincinnati. That county had a very powerful judge. He ran the court, administrative judge for many years. And this was a time when drugs were hitting that community very hard. He was tired of just locking people up, seeing them come back again. And he wanted a program. So he started asking around, where could he find a drug program that worked? They sent him down to Miami Beach. What better place to go for a drug program?

So he goes down to Miami and the judge down there tells him, “I have just the program for you: acupuncture.” And they trotted up some poor indigents who said, “Acupuncture changed my life.” And based on that anecdotal evidence, that judge came back to Ohio and he started an acupuncture program. Went something like this — the probation department ran it by the way, just so you know, right? Because probation works for the court in Ohio and the judge wanted this program, so this probation department ran it. It went something like this: they bought an old house down the street from the court and turned it into a day reporting center. Offenders had to report every morning at 8:30. First thing they did was pee in a cup. Second thing they did was get acupuncture, five spots in the ear. And then they spent the rest of the day in other kinds of programming: processing groups, drug education groups, 12-step groups, other things that don't work with offenders either, by the way.

So they did that for a while, spent a lot of money. Every time they did acupuncture it was 50 dollars. And these people were lined up every morning. Spent a lot of money and after about a year they called me up and they said, “Ed, we want you to come evaluate our program. We want to know if it works or not.” Well, let me be clear about something. In corrections, when somebody wants to know if a program works, they're not asking whether the offenders feel good about themselves, whether they like the program, whether they have more insight. They want to know, is it changing their behavior? Are they less likely to use drugs, steal, hurt people, commit crimes, right? In other words, are they reducing recidivism? I'm very clear about that. I'm in the crime business.

So my job was to evaluate the program. That's not hard, that's what I do. So I've got a whole bunch of drug-using offenders that go to the program. They had plenty of drug-using offenders that didn't go to the program. They're going to be my comparison group. Then they said to me, “We want to know if acupuncture works.” Now I had a problem, right? And this is a common problem. Because offenders weren't just getting acupuncture, they were getting all this other stuff. Think about it. Think about probation or parole. You get an offender on probation and parole in D.C. Do you drug test them? Yeah. Do you try to get them a job? Do you give them good advice? You can't give them enough of that good advice. Do you go over the conditions, make sure they understand … right? You do all of those things. And if the offender does well, you don't know why. You don't know if it's the drug test, or the fear of going to jail, or the advice you gave them. You just know that you threw everything at them and they did pretty well. Right? Well they wanted to know if acupuncture worked.

To answer that question I had to get them to allow me to do a random experiment. And they're tough to do. You know, it's hard to get folks to allow you to do them. But we randomly assigned offenders to one of three groups. Acupuncture: she'd come in in the morning; she'd pee in a cup; she'd get acupuncture; she'd get all that other good stuff. Placebo group: he'd come in in the morning; he'd pee in a cup; he got acupuncture in the wrong spot; [Control group] she'd come in in the morning; she'd pee in a cup; she wouldn't get acupuncture; she'd get all that other good stuff. Acupuncture, placebo, control, acupuncture, placebo … That's how you do random assignment. The only ones that knew who were in the acupuncture group and the placebo group was the secretary who kept the list and the acupuncturist. The PO's didn't know. The offenders didn't know.

And then we did what researchers do: we tracked the outcome, we looked at offenders, we controlled for differences. And at the end of the day, what did we find? Who did best? The control group. They had the lowest recidivism rate. [Laughter.] Now, it shouldn't surprise you because basically acupuncture is horse s***. [Laughter.] But … I thought about it over the years though. And I thought, “OK, there's no reason, we really think an offender, somebody that's addicted to drugs, long-term abuser is going to quit using drugs because we give them acupuncture. Do you really believe that? But to me it wasn't a question of why they didn't do better or didn't do as well. Why'd they actually do worse?

And I thought about this over the years and I really believe it's because those offenders thought they were getting some treatment that was going to fix them. They could do what they always did, hang around with the same folks, do the same things. It would be like me, you know, giving you a sugar pill or telling you, you can eat whatever you want under my diet you can eat as much or whatever you want, or I got an exercise program, I saw one in the airline newspaper the other day, right? Four minutes a day. Do you really believe you're going to get in great shape in four minutes a day? Do you really think you can eat anything you want and you're going to lose weight?

I mean, it's the same thing. We're telling offenders, “I got this great treatment for you. Here it is, the latest and greatest,” and they're doing what they always did. I'm telling you this because a lot of the things we've done in corrections, not only punishment-wise but in the name of treatment, doesn't work.

And here are a few of the things I've run across in the programs we've assessed. These are the so-called theories we've come across. “Offenders lack creativity” theory. I saw that once in California. It was a program for drug addicts. It was a community-based therapeutic program, they called it. It went something like this. Offenders would get up in the morning and they'd have a meeting and they'd do their chores and have lunch and then they spent the rest of the day doing art therapy: sculptures and pictures and really nice stuff. And I remember sitting down with the director and I said to her, “Why do you spend all of your programming time on art therapy.” And she said, “It reduces stress.” And I said, “You're correct.” I said, “You must have the most relaxed drug addicts in the state of California.” [Laughter.] I mean, think about it. What are they supposed to do when confronted with drugs on the streets? “Wait a minute I've got to draw you a picture. Don't go away. I'll be right back.”

“Offenders need back to nature” theory. We've seen that theory especially with juveniles. You know, the reason these kids are delinquent is they live in these tough urban neighborhoods and we just have to send them up into the country, you know, to chop wood and to grow their own food and learn some survival skills, you know, kind of approach, so when they escape from prison someday they'll be able to make it for a while out there on their own.

Of course, “need discipline and physical conditioning” theory. We've seen that in the form of boot camps. Take young men and get them in really good shape. I've never really understood that theory. I don't know why we would want offenders in really good shape. You think about it, we really ought to put them in couch potato programs, slow them down. I remember I was in a boot camp, I think — where's Gary? — I think Pennsylvania. I went to Pennsylvania's boot camp once and a young man, I asked him what he'd gotten out of the program and he leaned into my face and he said, “I'm in the best physical condition of my life, sir.” And I thought, now he can run me down and kick my a** even quicker.

“Change their diet” theory — that's a California one there too. [Laughter.] I saw this one once. It was actually a prison program, where when they acted up they made them wear a diaper and come to group dressed in a diaper. It's a shaming technique, right? Well, here's the problem. Does anyone here like to be shamed and humiliated? So what do you think you get when you shame and humiliate an antisocial offender? [Laughter.]

This was a mental health program in Ohio. We couldn't figure out what they were doing and they finally just said, “We want them to be happy.” So they have happy offenders in that program.

And my favorite — this was out of a New Jersey halfway house. “Male offenders need to get in touch with their feminine side” theory. This one went something like this: “The reason us men are aggressive brutes is we're just not in touch with our feminine side.” And so this program ran groups and they made them dress up in drag. They had to come to programming in drag. So if you're ever mugged in New Jersey by an ugly guy in a dress. [Laughter.] And people think I make this up. I don't make this up. I'm going to show you a few examples.

Here's one: “Dance Program Gets Juveniles Moving on the Right Track.” And there's what it says, “There in a small, secure concrete area young male offenders dance their way toward a new outlook on life.” The good news is, they're dancing to rappers like 2Pac, 50 Cent and R. Kelly [Laughter.] They've got some role models cooking there as well.

Here's one out of Tennessee: “Running Teaches Inmates the Value of Success.” I thought we didn't want them to run. [Laughter.]

This I like. This was actually a letter sent to Department of Corrections about drum circles and I'll read you. It said here, “She introduced the first drum circle in a New Zealand prison. She describes it as, ‘wow!'” That's the data they have right there, “Wow.” [Laughter.] The staff was amazed because most clients continued drumming for two hours without stopping to smoke. It doubles as a smoke cessation program — so you can kill two birds with one stone with that program.

And here they are. Right there, they're pounding their way out of delinquent behavior.

Let's see. What else? This was out of Texas: “Man Sentenced with Probation and Yoga.” Now that was for abusing his wife. And I like what the judge said. He said, “I thought about taking it myself but I got a pretty bad back.” [Laughter.]

And here they are right outside. Now they're meditating so I don't know what he's doing. But they're fixing themselves right out of court.

Here's another one: “Gardening Conquers All: How to Cut Your Jail Recidivism Rates in Half.” Fifty percent, mind you. I got a call once from a county — in fact, I think it was in Virginia — and they asked me, they said, “We're thinking of doing gardens for our offenders.” They said, “What do you think we'll get?” And I said, “Vegetables is what you'll get.”

And this, I like this is out of Canada. This is: “Dogsledding as Restorative Justice Method.” And here's what they said. “Exercising wilderness skills was seen as a way of rebuilding the perpetrator's self-esteem. So they're worried about how the offender feels about themselves. In Canada they're working on that.

These are all forms — oh, I have one more: “Hand Writing Formation Therapy Aims to Reform Juveniles in Texas.” For a fee you send this woman your delinquent's handwriting and she will fix them. She says it serves to change people's behavior by training them how to write in a different way. So these are all forms of quackery. They don't work. Unfortunately I'd love to tell you we still don't see examples, but we do. So that was my first lesson, some things don't work.

My second lesson is: if you want to reduce recidivism, you need to focus on offenders most likely to recidivate. This is known in the research as the risk principle. So if you have a group of offenders, you're all low risk. You're all low risk. Even if you got in trouble, you'd still be assessed as low risk. Why? You don't have criminal histories. You're not substance abusers. You have jobs. You're educated. Right? You don't have a lot of risk factors. Why would I put a lot of resources into trying to reduce your rate of recidivism? It's already going to be very low. Maybe 1 out of 10 of you will recidivate.

Rather, let's focus on groups where 60 percent recidivate, because that's where we're going to get the effect. But there's other parts of the risk principle that are important. One is that we've got to provide the most intensive intervention to higher risk. That's really a dosage issue. In the research, what do we see? The longer they're in treatment, the stronger the effects. Ah, but with a caveat. If you keep them too long, the effects go down. Well, that's not hard to understand either. People give up. If I put you in a program — if I had a great diet for you and I say, “I have a great diet but you have to follow it for three years.” Most of you'd say, “I can't do it for three years.” And so, I think that's one of the reasons we see these “lose weight in 12 hours” kind of ads, because they're appealing to, you know, the desire to do it quickly. But we can't do it that quickly. But it really becomes around dosage.

The third element of the risk principle is the one that we really have to be most concerned about, and that is, when you target lower risk offenders with intensive interventions, you can increase failure rates. It's well established. Bonta, Andrews, Gendreau and others first started articulating this principle about 20 years ago. We saw it in meta-analysis and in some small studies and we've seen it in some very large studies.

This study we did in 2002 in Ohio. At the time it was the largest study of community-based correctional facilities ever done. Basically it was a study of residential treatment facilities for offenders: halfway houses and what we call community-based correctional facilities. Director Wilkinson, who is the head of Ohio's Department of Rehabilitation and Correction at the time wanted — simple question: “Are they effective at reducing recidivism?” We're spending, you know $150 million a year, and he wanted to know, do they work or not? It's a simple question, we did the study. Large study, 13,000 offenders. I knew that you couldn't — if you're going to look at individuals that are coming out of prison, going into programs, you have to take in effect this risk principle. Think about it. Some people come out of prison, they've got a job waiting for them. They have a family waiting for them. Maybe they learned their lesson in prison. That's much different from someone coming out who thinks his problem's the government, who has no real job, not sure where he's going to be living day to day, who thinks his only mistake was getting caught. And so risk is very important.

And here's the mistake we've made as researchers. We often go in and do a study and what do we say? What do we say in the report? “The most successful offender was one who was employed, didn't use drugs, had a place to live, didn't have a long criminal history.” Like that's a big surprise to us. The mistake we're making is we're comparing low-risk offenders to high-risk offenders. That's a mistake. If you want to see a treatment effect, you have to compare high risk to high risk, low risk to low risk. And think about the notion that you can make people worse. If we were all healthy and we went to a doctor's office where everyone in the waiting room had H1N1, chances are a few of us are going to get sick from the experience. So we know that you can contaminate people. So we did this study and what did we find?

Well, these are the programs and that's not important to you, but these red bars, these numbers here are negative effects. That means, that's how much worse low-risk offenders did who were placed in these programs. Only a few programs showed positive effects with low-risk offenders.

When we look at the same programs and by high risk, now we see a number of programs with substantial effects, three up here around 30 percent. And by the way, we did have eight programs that did not work with anyone. They had no effect regardless of the risk. Anyone know what the scientific term for that is? S***y program. [Laughter.] That's the scientific term for that.

But here's where you see the risk principle right here. Let's just take this program, Mahoney County. This is where I grew up. Thirty-two percent effect with high-risk, increased recidivism 29 percent with low-risk offenders. Same program. As a researcher what it told me — by the way when you put their data together, you know what their average effect was? Zero. So had we not broken the data out by risk, we would have concluded that that program was not effective, when in fact that program was very effective with who? High risk. Who do we want to put in those kind of programs? Because we're spending 80 bucks a day on these kind of programs, we want to put in high risk.

Alright, so we did that study and there were some — all these folks said, “Well you didn't do this. You didn't do that,” and so forth.

We replicated the study; 2010, it just finished about a month ago, even larger. Twenty thousand offenders, 44 halfway houses, 20 community-based correctional facilities, parolees and probationers in the study. Match based on race, risk, sex, status, so there's a matching of offenders. Basically what we're comparing here is offenders who are placed in one of these intensive programs versus offenders who are just given supervision. So they're getting services. They're getting some treatment, but they're not getting it in the context of this 24-hour kind of program.

What did we find? Well, pretty similar. More programs, negative effects for low risk. Positive effects for high risk. But now we have a number of programs here in Ohio that had over 50 percent effects with high-risk offenders. And by the way, it was some of the same programs that were doing well in the 2002 study.

What did we find? Low risk, we increased recidivism by 3 percent; moderate, we reduced by 6; and high risk overall by 14 percent. And even 14 percent of 20 thousand offenders is not bad. So, who you target. If we're going to design these programs, we're going to spend all this money on them, we need to put people in that need them. As I told Director Wilkinson at the time, “Why would we spend $80 a day to increase recidivism rates?”

The third principle, or the third lesson I've learned is that everyone thinks they're an expert in criminal behavior. I'm not talking about you, some of you truly are experts, this is what you do. But think about it. Crime and criminal behavior is one of the few areas where almost everybody has an opinion. I'm sure some of you work with [inaudible] — I bet you have your family and friends tell you how you ought to fix these offenders. I bet you get a lot of unsolicited advice. Right? You do. They tell you we just need to do prayer in school or get tougher with them, or whatever it is. Everybody has their … “It's usually their mother's fault,” they say.

It came to me one day. I was flying, I was on a flight, I was flying to Boise and I was seated next to an elderly woman and she was one of these chatty types, you know, wouldn't take a hint, and she asked me what I did for a living. I made the mistake of telling her. And for four hours she told me how to solve the crime problem in America. Didn't get off in Salt Lake, stayed right on with me to Boise. [Laughter.] Now I just tell them I'm a proctologist, they leave me alone.

Understanding what we call risk factors is extremely important in designing effective correctional programming. Research by Andrews and Gendreau and others has led to the identification of some major risk factors. If you think about it, this work has been going on since criminology started. Lombroso was the first one. He was an Italian psychiatrist at the University of Pavia and he was obsessed with studying offenders. He personally studied 5,000 Italian convicts. And he wrote a book and in the book he wrote how one day they had killed an Italian bandit and brought the body to the prison. And Lombroso wrote that as he stood over this dead offender, it dawned on him after all these years of research that he could identify certain characteristics that distinguished the offender from the non-offender. And what were they? An enlarged jaw, cessual earlobes, tattoos, they were excessively hairy — have you ever seen an Italian that wasn't excessively hairy — acute eyesight, lack of foresight and a love of orgies — another Italian characteristic. [Laughter.] Lombroso went on to say, what, about a third of all offenders were what he called born offenders, born criminal. Very popular at the time. He was wrong. His work was flawed. Nobody believes that anymore. But why it's important is because he was the first person to try to study offenders scientifically.

Since Lombroso, there have been hundreds and hundreds of studies, and that's part of the problem. You start reading that literature, it'd take you years and by the time you got done you'd be just as confused as when you started, right? Because for every study that says, “this is a risk factor,” there's one that says it isn't. And so this work, the work of Andrews and Gendreau and others, is particularly, in terms of meta-analysis, have really helped us hone in on what are considered the major set.

Why's this important? Because if you want to change criminal behavior, you need to focus on the correlates of criminal behavior. I'm not talking about the reasons people become criminals. We have as many theories as we have offenders about that. I'm talking about somebody sitting across from you. Your job is to try to see they don't come back here again — what are you going to work on? There's things I can't change. I can't change that you were abused. I can't change your father abandoning you. I can't — certain things I can't deal with. I'm going to work on those things that I think impact your life now.

So, what are they? Well, things like antisocial, pro-criminal, attitudes, values, beliefs, cognitive emotional states. What are cognitive emotional states? Things like anger, rage, identity. If you identify yourself as a thug, how are you likely to behave and act? Attitudes, values, how you see the world, do you take responsibility? Do you think it's OK to take somebody's possessions? Is it alright to use drugs because who are you hurting but yourself? That's the rationale they often give. I interviewed an offender the other day on parole and I said, “Are you working?” He said, “No. I quit my job.” I said, “Why?” He said, “I wasn't getting enough hours.” I said, “How many are you getting now?” He said, “None.” I said, “You're moving backwards.” Most of us don't quit our job 'til we have another job lined up. Not him. He showed them. He went from 20 to 0. It's that kind of thinking that often leads to their behavior. They don't see the connection between their thoughts and their behavior.

Pro-criminal associates, we all know this. Your mothers all knew it. If you're a parent you know it. You worry about who your children hang around with. But it's not just knowing criminals, it's not having pro-social people in their lives. So part of what we've done in corrections, of course we do a lot of things wrong, we lock them up with all criminals, first of all. But even there, even when they're out, what do we do? “You can't hang around with this person. You can't go to this place.” We do all this risk management and we don't spend nearly enough time on risk reduction: identifying pro-social people in their lives, hooking them up with those people, whether it's an aunt or a grandma or a boss. And I could talk about how to do that but we don't have time. But that's extremely important area for offending.

Temperament. Social personality patterns. Things like weak socialization, impulsivity, adventurous, restless aggressiveness, egocentrism. It's a myth that low self esteem is associated with criminal conduct; it's high self esteem. But again, not by itself. Just because you're egocentric doesn't make you criminal. We'd have to lock up all the professors in this country, a bunch of judges too would go. But when you have that self-centeredness and you hang around with people that are criminal and you think what you do's OK and you don't have a job, now your risk starts to grow dramatically. And that's a point I want to make, an important point. You're not high risk because you have a risk factor. Offenders are high risk because they have multiple risk factors, and that has implications for programming. It's why we often don't get effects, because we take an offender who has all these risk factors and put them in a substance abuse program. But they got the same friends, they're hanging around the same places, they think the same thoughts. And we clean them up for 60 days and we send them back and we don't get the effects that we want.

It's simple to understand. If you don't think it's true, think about employment. For an offender in D.C., on your caseload, is being unemployed a risk factor? Absolutely. Is it a risk factor for all of you? If you lost your jobs, do you start selling drugs, mugging people, stealing cars? No. What would you do? You'd go get another job. Being unemployed for most people is not a major, major predictor for criminal behavior. Ah, but it is if you said things like, “I can make more money in a day than you make in a month.” “I'm not working for 10 bucks an hour.” “None of my friends are working and they're doing alright.” “I got somebody supporting me.” Now, being unemployed is a big risk factor because you have all those antisocial attributes and you have 40 hours a week to do nothing but get into trouble.

So understand the context of a risk factor. History is important. We all know history is a major predictor; what you did in the past is a great predictor of what you'll do in the future. But there are some caveats to it. One is you can't change it. And the second is you've got to have history before you can use it as a predictor. Think about that. I used to study habitual drunk drivers. You probably don't have any of them in Washington, but we had a lot of them in Ohio. And I'll let you in on a secret: you don't need a Ph.D. to know somebody with five DUIs is high risk for drinking and driving. How do you know it, by the way? Because they've got five DUIs, that's how you know it. But the same person came in the door, had one DUI, was high risk the minute they came in, but we don't know it, so we wait 'til they get the second, or the third, and the fourth, and then we say, “Oh my gosh, we have a high-risk drunk driver on our hands.” So the context of it, it's important — but if that's all you rely on is assessment, you're going to miss the mark.

Family factors, of course: low levels of caring and cohesiveness; outright neglect and abuse; levels of personal, financial, vocational and financial achievement; low levels of involvement in pro-social leisure activities; and substance abuse. These are what we call the major set. But I want to be clear. Most of us who study risk factors believe that these risk factors run through the big four. These are really the big four. If you can change somebody's attitudes and values and beliefs, if you can teach them coping skills and how to handle stressful situations, if you can work on — get them around pro-social people — the other areas come. The other areas often follow.

So I want to talk about a little study I did, it was a number of years ago, but it's a lesson I learned because it's really when this all started to come together for me. Anybody here sports fans? Any sports fans? I'm a sports fan. I'm a Buckeye. I went to Ohio State University and I teach at the University of Cincinnati and I've always been a sports fan. And Frank Cullen is one of my colleagues and Frank and I were always interested in athletes. And so a number of years ago we did a little study. We surveyed every Division I head football coach in America and we asked them about cheating. No one had ever done that before. And we got like a 75 percent response rate. And among the things the coaches told us was they believed that one of the reasons they had all this cheating problem was because they were forced to recruit all these poor kids, these kids from bad neighborhoods, these kids that grew up without any money, these kids that didn't have fathers in their lives. Their solution? Let us pay them a little bit. That's what they said to us. If we could pay them a little bit, cheating would go way down.

We wrote it up, we published it, we got a lot of attention. We were on all the radio shows and newspapers and we got a call from the NCAA. They called us and they said, “We liked your study. We want you to do another study. This time,” they said, “we want you to study athletes.” And so they gave us a small grant and Frank and I did a study of 2,000 Division I basketball and football players, male, randomly selected from every school in America. Every school has to turn in their roster. We took every fifth kid. We basically wanted to know, who were the cheaters?

Now, let me explain. Cheating athletics is not always the same as criminal behavior. So let's say my friend here, let's say he's a star halfback at the University of Maryland football team and I see him after a big game and he's with his friends at a restaurant and I say, “That's a great game you played today. Let me buy your dinner.” That's cheating. He's not entitled to a free meal from a booster because he's an athlete. The NCAA would rule that an infraction. The University of Maryland athletic department would tell him in no uncertain terms, “Don't let anybody see you doing that again. What's wrong with you?” [Laughter.]

All right. Let's say he's a star basketball player on the American University basketball team. They got a basketball team? All right, just checking. So he's the star guard, I see him before the big game. I say, “We got the money on the other guys tonight. Here's $25,000. I want you to throw up some bricks.” I know what you're thinking, that's all he's going to throw up anyway. But if he takes my money, that's a felony. He could go to prison for that. I'm telling you this because cheating varies, just like criminal behavior. We've got some stuff that's not that serious, and we've got stuff that's very serious. Cheating among athletes is the same way. We wanted to know who were the serious cheaters. This is what we found. Oh, by the way, it was called the gambling study because we're the guys who found that 25 percent gambled on games, but it wasn't a gambling study. It was a study of cheating, all types of cheating.

What did we find? Cheaters! The highly recruited, the superstars cheated more, not a myth. Well there's at least two reasons for that. One is opportunity. Nobody offers the third string punter a new car. But the other reason is because of what? Ego. Entitlement. “I'm the star. Why shouldn't I get something?”

Second, cheaters associated with fellow athletes who broke the rules, that saw nothing wrong with cheating. Cheaters hung around with cheaters. Third, they personally embraced values defining rule violations as acceptable. “Everybody does it. They'll take care of it. Coach gets a lot. School gets a lot. Why shouldn't I get something? I'm turning pro anyway.” Attitudes, values, beliefs. Fourth, did not [inaudible] with their parents or coaches. Remember, for many of these kids, coaches are like surrogate parents. Not these ones. So here you have family and isolation from more pro-social people. And finally, they had a history of delinquent behavior. Those are major risk factors right there. Now you're not going to see substance abuse. They're all in college so they're all drinking too much. You're not going to see education, they're all in college. But here what you see are the big four in a non-offender sample, same as we see with offenders.

By the way, what wasn't significant? Remember what those coaches told us? “It's those poor kids.” Here's what they were really telling us. “It's those poor black kids we recruit.” That's what they were really saying. Guess what? Our sample was 50 percent black, 50 percent white. Race was not a predictor. It did not matter. Economic deprivation did not matter. Coming from a poor background did not matter. Having money in college — how many of you had money when you went to college? — it didn't matter. Organizational contacts. The NCAA had what they called the “bad apple” theory. Renegade programs. Renegade coaches. They said, “If we could just clean up these coaches, we'll clean it up.” No. Did not matter. We looked at region, school. Didn't matter. I'm a Buckeye. I went to Ohio State. When I did this study, Michigan had beaten us five years straight. I'd have turned their a***s in if I saw they were cheating.

[Laughter.]

Latessa: Finally one important [inaudible]: threats and sanctions. The certainty and severity of punishment for violating the rules was unrelated to infraction. Oh, you can scare the low-risk kids. But here's the problem: they weren't the cheaters. I can scare you. But the ones we want to scare most, punishment isn't — they're immune to punishment for a bunch of reasons.

Number four, doing things well, make a difference. This is true in almost anything we do in life. If you don't do it well, don't expect good results. And it's certainly true in corrections. I first started doing this work about 20 years ago. We had a study in Ohio of day reporting centers. The state wanted to know if they worked or not. Obvious question. But they also wanted, they had an interesting question there. They said, “If they work, we want to know why.” So we had five of these around the state and they were all getting the same offenders and the same money and the state says, “You know, tell us why one's more effective than the other.” And so I had to start to figure out, how do you get inside of a program? How do you look at those things like assessment and the programming and the staff, the kind of things that most of you would tell me, if I asked you, “What do you want? What would you need to design a good correctional program?” That's what you would tell me. You'd say, “I want a qualified staff. I want a curriculum. I want resources. I want …” and that's what we look at.

When we look at the data, this is out of Washington State, Steve Ose's work. What did they find? Even with evidence-based programs, things like functional family therapy and aggression replacement therapy. Very significant effects, but only when competently delivered. If they're not competently delivered, they actually produce negative effects.

This is why program integrity is so important, because people will go out and pick up one of these blueprint programs or one of these curriculums and they wonder, “Why aren't we getting the same effect?”

I've done several large studies. Each of them has found a strong relationship between program integrity and recidivism. The higher the integrity score, the lower the recidivism rate.

In our halfway house study in Ohio and our residential programs study, poorly designed, poorly implemented programs increased recidivism 19 percent on average. The highest scoring programs reduced recidivism on average 22 percent. That's a 40 percent swing. And guess what, my guess is you start putting them together, what happens? It looks like the programs aren't working. It's another reason you have to break programs out when you do the research.

We did a study of community supervision programs. Things like day reporting centers, ISP programs and so forth, electronic monitoring. Guess what? Type of programming did not matter. What mattered was the principles as well as integrity. Poorly designed programs; well designed programs. Thirty percent swing, and that study had 13,000 offenders in it as well.

Juveniles. Did a big study in Ohio of juvenile programs and of course a lot of juvenile programs are designed for lower risk kids — they want to do prevention and other kind of programming. And that's fine. And guess what we found. This was their program score. So the white is the higher program score, and at every risk level, lower recidivism. Poorly designed programs; better designed programs. So quality makes a difference when you design and operate correctional programs.

And the final lesson is that we can change offender behavior, we just need to go about it the right way.

And I've drawn on this from some personal experience in raising these four semi-delinquent youth. These are my children and they're all grown up now.

But I learned a lesson with this one. [Laughter.] This was the youngest of the family, Allison. She always had a temper on her and being the youngest, she kind of got picked on. Things flow downhill. This one was very good at picking on her. And so one day when she was probably not far from this age something happened and she got mad at him and she threw the portable phone and bounced it off his head and was standing over him threatening him and I caught her. And I said, “Allison! What are you doing?” And she's looking at him and I said, “How many times have I told you, you are not to take things into your own hands and hurt your brother?” And she said, “I don't know, I thought you were keeping track.” I said, “That's it Allison. Into the corner.”

Now I'm punishing her. The reason I'm punishing her is I want to extinguish the antisocial behavior. That's the purpose here. I want to calm her down. She's in a cognitive emotional state of rage. She is madder than hell. She's not taking it any more. I've got to deal with that so I've put her in the corner. And some lessons in how to punish effectively too by the way, there's rules to punishment. I don't have time to go into them but one is, how long? Now, I could've put her in the corner for a half hour. What would've happened? She'd have forgot why she was there. It's too long. I did forget her once, I was watching television. But she's OK now. So say she was five or six, I put her in the corner for five or six minutes because that's what child psychologists tell us: about a minute for every year they are. And I turned on the timer. Six minutes, five minutes whatever.

So first minute or two she's still angry. She's looking at her brother and she's looking at me. “When I'm done with him I'm coming after you.” That's what she's thinking because she's in that emotional state. About minute three or four, now she's calming down but she's pouting. Now it's the blaming others kind of thing. “I wish I was dead.” That would fix you. You know the big lip comes out so she's in that blame the victim kind of thing. She's the victim here. About minute five she's calmed down. She says, “Dad, I'm ready to get out of the corner.” Now, the punishment's worked. I've extinguished the behavior. That was the goal of the punishment. But I haven't replaced it yet. So I say to her, “Allison, wait ‘til the timer goes off and then we're going to talk.”

The timer goes off. She comes out. She looks at me. She says, “I'm sorry. I won't do it again.” I said, “No, no Allison. Sit down. We're going to talk about what happened.” I said, “Tell me what happened. Tell me what you did.” She said, “Michael did this.” I said, “That's criminal thinking. We're not talking about Michael. We're talking about what you did.” She kind of looked at me funny. She said, “Well. I got mad. I hit him with the phone.” I said to her, “Allison, what are you supposed to do? What have we told you to do over and over again when he does whatever it is he does to you?” And she looked at me just as bold as could be and she says, “We're supposed to come get you or mom. I'm supposed to come get you or mom.” And I realized right then the mistake I'd been making. She knew the words, but not the music. She's not stupid. You tell her 50 times, “Don't you do that. You come get me and mom. We'll take care of it.” She's learned that. But let's play it out a little differently. There she is. Her brother does something to her, aggravates her, upsets her. She lets him have it. You think she feels good when she lets him have it? She feels good. She took care of him. She showed him. Does she get caught all the time? No. And what's she thinking, “I'll do my five minutes in the corner.” [Laughter.] Now she's five, she's not thinking that but that's what offenders are thinking. “I'll do a nickel. I'll do five years.”

So she's thinking, “I took care of it myself. I felt pretty good about it. I showed him who the boss was. He's not going to be doing that for a while. He'll get a knot on his head.” She only gets caught once in a while. She'll take the punishment. So I said, “Look obviously.” I'm thinking to myself, “She doesn't know what to do. She's heard me, but she doesn't know what to do. So I looked at her and I said, “Allison, do you like getting into trouble?” She said, “No. I don't like getting into trouble, obviously.” I said, “How about if I could show you a way to handle that situation where you wouldn't get into trouble.” She said, “You could do that?” I said, “I can, but you've got to make me two promises, only two. You have to try it, and you have to let me know if it works or not. You think you could do that?” She said, “I could do that.” So I said, “Alright, show me what he did,” and she did. And we practiced and we rehearsed how she could handle that. I taught her some skills. And she had it.

And I said, “Now what if he does this to you?” because Michael's going to escalate it, right? Think about it. You're on probation. You know you shouldn't go out to this party because there's going to be drugs, there's going to be trouble. Your friends are there. They say, “C'mon. We're gonna go. We're gonna have a ball.” Putting a lot of pressure on you. Cognitively you get it. The problem is you don't know how to get out of that situation. And maybe you try the one thing you were taught and it doesn't work. Now what are you going to do?

So I knew I had to teach her more. So I said, “How about if he comes at you like this?” And she says, “Oh, I'll really let him have it.” And I said, “Well let me show you another way,” because you can't always use one way to get out of a situation. Most of us have a lot of skills to get out of situations we get in. Offenders do what? They do what they always do. And so we practiced that. And I said, “Now, remember what I taught you. Remember what I told you. You have to try it, and you have to tell me if it works.” So sure enough he did something to her in a few days and she did what I taught her and it worked, and she came back and she was very pleased with herself. And she said, “Dad, you should have seen the look on his face.” I said, “That's great. Now I've got to go straighten his a** out.” [Laughter.]

My point of it is though, that I interview offenders all the time coming out of prison, men and women, all the time, and I'll ask them. Here's what they'll say to me. “I've learned my lesson. I'm never coming back again. I hate this place. I'm going to go out and I'm going to do this and this and this and it's going to be great.” And I'll say to them, “What are you going to do when you get into this situation?” Whatever it is. And they'll look at me and they'll say, “Well, that isn't going to happen.” And I'll go, “Why?” “Well, my friends know I'm going straight.” And I'm like, “Oh, how do they know? Is it on the Internet or something?” I mean, what it tells me immediately is, they've not practiced, they've not rehearsed, they don't have a plan; no one's taught them the skills they need. We spend all of our time trying to motivate them or threaten them or scare them and we don't teach them how. Now they won't all do it, they won't all use the skill, but that doesn't mean we shouldn't be teaching them those skills when they come out.

So my last lesson is really that you've got to use behavioral approaches and that means focus on current risk factors. I go to programs all the time and they want to talk about the past and they want to talk about, you know, events that occurred 20 years ago. They want to talk about the criminal behavior. You can't change that. You want to focus on the current risk factors. You're using drugs. You're not going to work. You think it's OK to steal. You're hanging around with the wrong crowd. Those are current risk factors. Second, of course, our interventions need to be action oriented. And by that I mean, we need to teach them how. Not simply talk to them. People don't change their behavior because they hear speeches. It's not a very effective way to change somebody's behavior. You have to model it: social learning, cognitive behavioral interventions.

Think about yourselves. Think about why are you doing what you're doing? Who influenced you? Who'd you learn from? Who'd you want to be like? Social learning is the strongest theory we have about who we are. It's how we acquire our attitudes, our values, our beliefs, from those around us. You don't believe me? How many of you have children? How many of you turned into your parents when you had children? You woke up one day, you're your mother. The last person you were going to be at 15 was your mother or your father. I have a — Ali said to me once, when she's a parent she's going to let her kids do whatever they want. I said, “You go right ahead, just be home at 10 o'clock.” I've turned into my father. I say things I swore I'd never say. “You born in a barn?” “You think money grows on trees?” “I'll give you something to cry about.” That was always my favorite. My son one day, I said to him, “Michael, do I look like Rockefeller?” He said, “Who the hell is Rockefeller, Dad?” I said, “I'm sorry, son. Do I look like Bill Gates?”

But that's social learning. But you can't leave it to chance. It's structured social learning. It's where you take a probation officer, sometimes that's the only pro-social model that offender has. They have to very structurally teach that offender new skills and work with them, not simply meet with them for six minutes and go over the conditions every time. If we're going to be agents of change, we have to go about it in a systematic kind of way.

So I think I'm about run out of time. I want to thank you all and I will answer any questions that you might have.

[Applause.]

Phelan Wyrick: Hi. Thank you very much. You sort of came across as the Lewis Black of criminology.

Latessa: I've heard that before. And by the way, I didn't think any of it — where I come from, the words I didn't use aren't bad words. [Laughter.]

Wyrick: And he's local to this area too, so we're OK with that. So, but I found myself just, even as I was agreeing with you and laughing along as you were talking about the different theories that have been used and the different approaches, I found myself just a little uneasy at that early part when you were talking about the various things, the dance and the acupuncture and all the various theories that were used. And the reason I was a little uneasy is because I've heard folks talk about, you know, well you hook them, you hook their interests with these various activities, but then you come in and you provide the kinds of, you know, maybe cognitive behavioral or whatever types of supports. So I just thought I'd throw that out and see your reaction to that.

Latessa: Ah, I've heard that often, and I think readiness to change, commitment to change is certainly a barrier that offenders have. Most of us have it, by the way. Most of us don't wake up some day and say, “I'm just going to change my life.” I mean, somebody's pushing you, somebody's prodding you. And so, the idea that somehow, getting you in those activities is going to engage you and then you're going to move forward. If I thought that's how they were systematically doing it — there are programs. I often advocate that programs develop pretreatment programs. But again, very systematic, in which you take those highly resistant offenders and you use motivational interviewing techniques. Texas Christian University has actually developed some curriculum and I've seen it done with some hard core guys, in which they start to talk about where they're at and where they want to be, and getting them ready for programming. I would advocate that approach than this.

I have no problem with activities. Offenders need activities, especially if you're working with kids in prison and what have you. But I think in corrections, I tell the folks I work with, “Distinguish between your core correctional programs and your activities. You want to do art, you want to do choir, you want to do rec, that's fine. People need to express themselves. But don't sell it as a program that is going to reduce risk factors.” Because, you're right, somebody may connect and have a great voice and go on, but that's not how we're going to bring about change in most offenders in corrections. So, I appreciate those comments, but I think we have to think about how we're going to do those things within context. If not, you just have offenders that can dance. Probably quite a few can dance now, so I'm not sure you're getting …

Julius Dupree: Hello, I'm Julius Dupree, I'm with the Bureau of Justice Assistance. I just wanted to go back to the piece where you were talking about where to focus the programming resources, low risk versus moderate risk versus high risk. I guess what I was really thinking about is that there is some research, I guess it suggests that some of the extremely high-risk offenders, you know, like your Henry Lee Lucas, your Ted Bundys, may not, may be — programming may not be efficacious for those folks, and so I just wanted to kind of get your thoughts on that.

Latessa: Yes, yes. You're absolutely right. And if you listen carefully, I talk about higher risk; I don't talk about the highest risk. Risk is really a continuum when we measure it, right? It's a fairly normal curve: we've got some low-risk offenders, we've got some very, very high risk, we've got a whole bunch in the middle, moderate and so forth. And for the offenders that would have a diagnosed antisocial personality disorder, the psychopathic offenders, we really don't have any known treatment. There is some evidence, by the way, that putting them into programs actually can make them worse because they, I mean, think about it, they're often bright, manipulative, so they actually get more skill at using folks. They can be very disruptive to a program. I'll be honest, it's one of the reasons when we look at correctional programs, one of the first things we look at is, what are the screening criteria for the program? I go to programs all the time that have very loose screening.

I'll give you an example: therapeutic communities in prisons. Oftentimes they basically take anyone that wants to be there, and it's an innate driven model. So I go in, and in five minutes I see inmates running the groups. And who's running the groups? The psychopaths. I mean the bright, smart, manipulative offenders. They're telling all the others what they need to do. So, your point is very well taken. Fortunately it's a fairly small, now you're going to have a lot more in a maximum security prison than you are … I mean, there's a screening in prison, but in the community they aren't. And they're not all antisocial, not all psychopaths. We've got some selling cars and insurance and things like that; Wall Street had a whole bunch. So they're not all, but there is that small piece for which we have really not designed any, and I don't know if we will. I'm not working on it, so I don't know if we will.

Bob Ries: All right, my name is Bob Ries with BJA. My question for you is: in your opinion — to what degree do you feel the adequate sharing of information or the lack thereof between corrections and re-entry service providers impacts the success of re-entry programs and recidivism?

Latessa: It's a good question. My experience is there's very little sharing of information at any level. People are put on probation and then sent to a referral agency, and they send them no information. I mean, even that quick connection often doesn't, we don't do it. Part of it I think is, at least in many states I work, the institutional folks don't do a particularly good job of assessing where someone's at before they come out of prison. They don't particularly do it. They can say, “Oh, he went to this program and that program,” but they can't really tell you what he learned. The discharge plans don't have the information and the detail. I think it's critical, if you're going to design a re-entry model, that we figure out a way for that information to flow.

I'll give you an example about institutions which are a little different. We recently developed an assessment system in Ohio, and it has a number of tools: community supervision, pretrial and so forth; nothing unusual about those tools. But the one I was most interested in was the re-entry tool, and here's why. Most systems assess offenders when they come in the door and most risk factors are still fairly accurate. Where were you living? Were you using drugs? Were you working? And all that, and you can assess the guy. Then they bring him into prison for five years, four years. And then the assessment they do is usually what? Reclassification based on institutional behavior, because they're moving him down.

Well we went in and we assessed men and women who had done at least three years in prison on the typical risk areas. What we found was only three domains predicted failure when they were getting ready to come out: criminal history, social support and social capital, and criminal attitudes and values. All the other domains that are typically in play when they come in disappear. Well, you think about it, the guy's in prison for three years. You're asking him where he lived? You're asking him who he hung around with? I mean, he's in an institution. You ask him what drugs he used? I mean, even if he had some, he didn't have that street corner access that they have, so a lot of things disappear. That kind of information could help you then when you decide what kind of programming or what level of supervision are we going to get. But I've rarely seen that kind of — and we're just starting to do it now in Ohio, so we have no data. But that kind of assessment, I think, has to continue as they come out because when they come out, now other risk factors start to come back into play, because now you make choices. Are you going to work that day? Are you going to drink? Are you going to hang out? And so you have to have an assessment and information that moves along the continuum, but we don't do a very good job and I do think it contributes quite a bit to our failure to do re-entry well. Long answer for a short question, sorry.

Phyllis Turner-Lawrence: My name is Phyllis Turner-Lawrence. I'm an independent consultant in restorative justice and victim issues. I'd like to ask, and I don't know if you've studied this, but I know that Ohio for one is one of the states that early on started with victim-offender dialogue programs with victims who wanted to volunteer, who wanted to go in and have dialogue with the person who committed the offense. But not just that realm, but also developing empathy and understanding of their own victimization and how they're victimizing other people. Where do you see that fitting in as part of programming, or being important, not important? Thank you.

Latessa: Yeah, lack of empathy, that cold heartedness is part of that personality pattern that was up there. It is a risk factor. The problem is, it's very difficult to address. What we have found is that there's two types of empathy. One is cognitive empathy and the other's emotional empathy. Cognitive is: I understand I hurt you and I hurt other people. Emotional is actually feeling their pain. And there's an instrument that kind of measures empathy. And so I've had some students do some studies with sex offenders, for example, on it and what they find is they can change the cognitive; the emotional, it's much more difficult. And I'm not an expert on how you would address it, but we try everything from give them a pet in prison, everything else, and you know, horse therapy (and there's another word that I would use with that kind of therapy). But I don't know if we've had a lot of success in targeting it; but again I'm not an expert in it. But there are ways to measure it, and so if I were designing a program to try to address empathy, I would probably measure it and see how much change I got and see what worked and what didn't work. It is definitely within that domain of personality, empathy.

Robin Delany-Shabazz: Hi Dr. Latessa. I'm Robin Delany-Shabazz with the Office of Juvenile Justice and Delinquency Prevention. Early on you talked a little bit about, with the high-risk cheating athletes, that the threat of deterrence did not work. So can you talk a little bit about how that applies in juvenile justice and criminal justice?

Latessa: Well, the idea of deterrence, there are some assumptions behind it. One is that you have a rational actor, that they assess the risk, that they think about what the odds of getting caught, they make some informed decision. The problem is, most street-level offenders failed at school, fail at work, fail in relationships, hang around with people just like themselves, abuse drug and alcohol. They're not very rational. So I'm not, you know, deterrence has a place, and certainly some of us have … we like to say we don't do it because … that's really not why most of us don't commit serious crimes. It's because of our attitudes, our values, our beliefs and other things. But, a best example was, I remember when the Enron folks, you know, were convicted, and they trotted them all up and they said, “This is what happens to you.” And clearly it had an effect on Wall Street. You see what deterrence it had on them, so with kids in particular, I've raised four teenagers. I've never seen one yet think about, you know, the consequences of a behavior.

So the idea that we're going to scare or threaten or punish them into change is simply not supported by the data. And I'm an empiricist. If punishment was the most effective way to change people's behavior, I'd be telling you how to punish them more effectively. But it's simply, there are ways within a behavioral program, for example, within a behavioral program I've got to, what, I've got to reinforce your behavior. That means if you're doing the right thing, I make sure you know you're doing the right thing. But if you're doing the wrong thing, I've got to stop that behavior, so punishment can play a role. But then I have to teach people how to do it effectively. And that means that you can't escape the punishment. It has to be, one of the most effective things like response cost or the most effective type of punisher, so within the context of programming, you teach some of those skills. But just this notion that general deterrence, we're going to scare or threaten kids, forget about it. Unless we see data to the contrary; I'm not seeing much data. Think about yourselves. You get a speeding ticket, you slow down for maybe 20 miles. I mean, you didn't even pay the ticket yet and you're speeding again. So I'm not sure how much effect we get from deterrence.

Debra Whitcomb: I'm Dr. Whitcomb with the Justice Management Institute, and we are working with BJA on a training program we call “Smarter Sentencing,” and so we bring a lot of the content that you shared with us to teams of prosecutors, defense attorneys, judges and community corrections folks around the country. From our perspective, prosecutors are really key. The front end of the system, even before the judges, they're the ones who are negotiating the pleas. Do you have any tips for us on how best to persuade them of the value of this kind of research for what they do?

Latessa: That's a great question. Two things: I recently spoke to the Ohio Prosecutors Association annual meeting and it was a … I was the last speaker Friday afternoon, 3:00. I mean, this is like the death knell. And I'll tell you, they all stayed. I've been invited now, I've done a couple of different workshops now around Ohio by the prosecutors, but they were very interested in not only this, but we talked about the new assessment tools in Ohio and how they would be used. So there was a lot of interest on the part of prosecutors because I think they understand that low-risk people don't need these kind of programs.

The state that's done the most in my opinion in working with judges and prosecutors has been Illinois. I've been doing work there for a long time. Because of the way they're organized, their administrative office of the courts handles probation, and they bought very early into evidence-based practices and every year I go and do a series of workshops around the state with prosecutors and judges. They actually do sessions where they teach them how to interpret the assessment information. They work hard on it because one of the challenges is how do you, with plea bargains forming most of our decisions, and they make those independent of assessment and all this other, so how do you work that in? And there's a number of judges, Judge Warren who does a lot of this work. I have a judge in Indiana who often has done it with me and they talk about it, especially when a judge talks about it, they really pay attention. So I would look at Illinois and what they're doing to educated judges. I think they probably have the best model.

Rose: I'm going to just jump in here. We just have a very few minutes here. I believe you have the next question, so I'm going to invite you to ask that question, and then I'm going to save these questions for Dr. Latessa after the lecture so that we can end on time. OK, go right ahead.

Lia Gormsen: I'm Lia Gormsen, with the American Correctional Association, and I'm wondering, Dr. Latessa, if you've found that those kind of top four risk factors, and particularly the personality profile, apply equally to male and female offenders.

Latessa: There's some differences. For example, weak verbal skills is often associated with men. They can't articulate their needs very well, so that's more common in men. But lack of coping skills, probably just as high in women. I've never really looked at all those areas specifically. It is a good question though, and I will tell you that I think pathways to criminal behavior I think are different for women and men. I think there are some differences within the domains. But that said, the data, and I've looked at thousands of cases of LSI data and other data, the basic domains are the same for men and for women. You think about it, right? Women, if they're hanging around with bad folks, who are they? Bad men. So there's a difference, but it's still relationships, it's still antisocial peers. With men they're hanging around with bad men too, so there are some nuances. But when you look at the data, LSI predicts just as well for women. Cognitive behavioral studies I've seen do as well if not better for women. So I think if you go with the evidence, what would I tell you? Focus on the, assess them. Focus on the domains. Do behavioral interventions. Do it well. I wouldn't put men and women together in groups. Our studies show that's not very effective. I don't know why. I think they're thinking of something else or something. But the studies we've done show you can get good effects.

Gormsen: Thank you.

Latessa: That's it.

Rose: Please join me in giving a warm round of applause for Dr. Latessa. Thank you so much.

[Applause.]