National Institute of Corrections and DC Pretrial-Measuring What Matters-DC Public Safety

See for our television shows, blog and transcripts.

Radio Program available at

We welcome your comments or suggestions at or at Twitter at

[Audio Begins]

Len Sipes:  From the nation’s capital, this is DC Public Safety.  I’m your host Leonard Sipes back after a three-month hiatus due to flipping a motorcycle and being injured, and finally back to doing weekly radio programs.  For those of you who have been kind enough to inquire, “Where you been, Len?”  Well, that’s where I’ve been.  I’ve been laid up telecommuting and away from the microphones, but we do have a really interesting show today, ladies and gentlemen, on pretrial.  The whole concept is measuring what pretrial does, and so pretrial agencies, both in Washington DC and throughout the country can do a better job of pretrial supervision.  We have two experts with us today.  Spurgeon Kennedy, he is the director of research analysis and development of the Pretrial Services Agency for the District of Columbia.  WWW.DCPSA.GOV .  I’ll be giving those website addresses throughout the program.  Lori Eville, the correctional programs specialist for the National Institute of Corrections, and the National Institute of Corrections is part of the Bureau of Prisons, US Department of Justice, WWW.NICIC.GOV , and to Spurgeon and to Lori, welcome to DC Public Safety.

Spurgeon Kennedy:  Thank you Len.

Lori Eville: Yes, thanks for having us.

Spurgeon Kennedy:  And welcome back.

Len Sipes:  Thank very much.  Spurgeon, the first question is going to go to you.  There’s a lot of people out there in the criminal justice system who listen to this program on a regular basis, and they’re going to know exactly what pretrial is, and they’re going to know exactly what jails are, but there are people from mayors offices or citizens or community organizations that listen to this program and they’re not quite sure what we mean by pretrial.  What is pretrial supervision?

Spurgeon Kennedy:  Well, simple answer is that pretrial services agencies help their local jurisdictions meet the requirements their state bail laws.  Most bail laws across the country have a presumption of release for most defendants except those charged with capital offenses or very serious crimes.  Pretrial agencies across the country help assess the various levels of risk that these defendants present and offer appropriate releases supervisions conditions that meet those risk levels.

Len Sipes:  Now you do know because I’ve been to community meetings.  You’ve been to community meetings throughout our career.  You go into community meetings and somebody says, “He got arrested, and he was back on the street in six hours, and that’s wrong.  You know, he did a crime, and he’s back in the community.  What gives with that?”  And the point in all of this is it not that people who are arrested are considered innocent until proven guilty.  Until a judge or a jury finds them guilty and sentence has been pronounced, so up until that point, technically, that person is a “innocent” person, and somebody’s got to make an assessment based upon two things, whether or not he or she is going to return to trial on their own with supervision or with bail or with some other arrangement, or B, they’re a risk to public safety.  So do I have – is that summation correct?

Spurgeon Kennedy:  Yeah, you’re absolutely right.  There’s a huge difference between a convicted defendant and a pretrial defendant, and what we deal with on the pretrial level are those persons who are still considered to be innocent or presumed innocent until that point of disposition.  What we’ve found over the last decade, especially when you look at the risk of defendants who are released back into the community, and certainly that’s a legitimate concern.  Public safety is always a concern for anybody in the criminal justice business, but what we have found is that defendants by and large who are released pretrial present a low to medium level of risk of failure to appear or to commit another offense while on supervision, so by identifying this risk and offering supervision levels.  Pretrial agencies across the country actually help their jurisdictions manage that risk, and to make sure that defendants who are released are not presenting an overly concern to public safety.

Len Sipes:  And the last question before going over to Lori is the sense that people need to understand, I think, that it is literally mathematically impossible.  There are tens of millions of arrests in the United States every year.  There aren’t tens of millions of jail beds, so it is, even if we didn’t have the presumption of innocence.  Even if we didn’t have a presumption of release on a pretrial basis, it’s mathematically impossible to keep all people arrested behind bars until trial, correct.

Spurgeon Kennedy:  You’re absolutely right.  In 2010, the National Center for State Courts estimated there were 21 million filings of felony and misdemeanor cases across the country at local courts.  Our jail population cannot support that.

Len Sipes:  No, no.

Spurgeon Kennedy:  And the risk, again that most pretrial defendants present, and the laws that we live under really don’t allow us to consider jail first.  So again, we need a system that identifies those defendants appropriate for release according to the law and those that should be detained pretrial, and that really is the basis of what pretrial programs –

Len Sipes:  And the final thing I did want to mention in terms of the Pretrial Services Agency for the District of Columbia, one of the things that I noticed years ago, is that you have an 88 percent return rate.  The overwhelming majority of people on your case loads show up for trial.

Spurgeon Kennedy:  True.

Len Sipes:  And it’s a much higher return rate than the national average as measured by The Bureau of Justice Statistics, so first of all congratulations, on doing a very good job.

Spurgeon Kennedy:  Well, thank you.

Len Sipes:  Lori Eville, correctional program specialist for the National Institute of Corrections.  What is the National Institute of Corrections, Lori?

Lori Eville:  Well, the National Institute of Corrections, as you said, is within the Department of Justice, and it’s primary function is to the provide support to federal, state, and local correctional agencies throughout the United States.

Len Sipes:  Uh-huh.

Lori Eville:  We do that through technical assistance, through education, an information center in which people can go to a vast library to get information that is particular to issues that they’re dealing with, and NIC is also a leader in developing policy and practices, and looking sort of a step ahead.

Len Sipes:  Uh-huh.

Lori Eville:  Of where corrections agencies and criminal justice agencies need to move or are moving, and give them support, and it’s really one of the primary purposes why NIC is interested in pretrial services because it is a distinct function within a criminal justice system, but it’s an essential contributor to the efficiency of a criminal justice system, as you said.  Not a fiscal efficiency, but as well as a case processing efficiency.

Len Sipes:  Well, we do have to talk about the fiscal part of it because the principle subject of my reading criminal justice literature newspaper articles and radio reports, television reports over the last two years is the fiscals, dire fiscal shape that states and localities find themselves in, so, you know, you just can’t lock up everybody.  It is just mathematically impossible.  It is fiscally impossible to lock up everybody, so one of the things that you did in terms of this report, and the report is measuring what matters from the National Institute of Corrections Outcomes and Performance Measures for the Pretrial Services Field.  What you’re trying to do is to frame this whole concept of pretrial, to get everybody to measure what it is they’re doing so they can figure out how they’re doing and how they can improve.

Lori Eville:  That’s correct, and it’s also distinguishing pretrial outcomes and measures from other criminal justices functions such as probation, jails, and we don’t have a document like this.  We have these established measures for other criminal justice functions, but we don’t in pretrial and as front end of the system, functions are becoming more evident in their need.  That’s why we sought to develop a consistent set of recommended outcomes and measures along with the definitions so we can get to comparing different jurisdictional functioning.  Looking at nation averages, and then also focusing on those things that pretrial services should produce and that is appearance rates.  That’s what a primary function was, a pretrial services should be, is that they have good appearance rates to court, and that they have good public safety records.

Len Sipes:  Okay.  Getting them to court.  That’s the bottom line in protecting public safety.

Lori Eville:  That’s the bottom line in pretrial.

Len Sipes:  Before you go on, I do want to mention, the National Institute of Corrections is the place where the rest of us go to get information.

Lori Eville:  Yes.

Len Sipes:  I mean that’s the, you know, I’ve been involved in the criminal justice system for 42 years.  I’ve been doing public relations for 30.  I’ve been doing corrections for 20, and it was interesting that when I first became a public affairs officer for a large agency in the State of Maryland that had corrections he as well as laws enforcement, as well as the fire Marshall’s office, as well as lots of other agencies, I realized that I didn’t have any formal training as a public affairs director, and the first thing NIC did was to put me in Boulder, Colorado for a couple weeks of training.  All throughout my career, when we said we needed research, or we needed funding, or we needed to look at this issue, we picked up the phone and called the National Institute of Corrections, so I want to thank you in terms of my own experience, and my own criminal justice career.  I’m not quite sure how, if a lot of people know the National Institute of Corrections.  WWW.NICIC.GOV, but a certainly thank you to NIC for all the help that you’ve given me, and the agencies that I’ve represented throughout my career.  Okay.  So we’re going to get back to larger issues in terms of measuring what matters.  The average person is going to sit back, and he’s going to listen to this program, or she’s going to listen to this program, and they’re going to say, what are we talking about?  This is 2011 we’re, you know, I thought you guys were doing this for decades.  We in the criminal justice system, Spurgeon, don’t do a very good job of measuring things, do we?  I mean most of the criminal justice agencies have a hard time coming to grips with measurement, correct?

Spurgeon Kennedy:  It is, unless you’re a nerd like me, talking about numbers and measures will put you to sleep as quickly as anything.  Here’s the thing though, and you’re right, we’ve been very slow to come to the table.  Businesses across the world have used outcome and performance measures almost since the mid 50s.  This is something that just became popular within criminal justice in the mid 1990’s.  1995, in fact, the American Probation and Parole Association along with the National Institute of Justice put out what I think was the first article about outcome and performance measures or if the probation field.

Len Sipes:  Uh-huh.

Spurgeon Kennedy:  That was followed by a larger publications looking at outcome and performance measures for the entire criminal justice system that NIJ did.  So it’s really been since the mid 90s that this whole ideas of measuring what you do, especially measuring against what you have out there is as your mission and your goal has become popular within criminal justice agencies.  Pretrial programs, as Lori mentioned, are just getting into the idea across the country, of measuring their performance and knowing what that performance ought to be.

Len Sipes:  Before the program, we sort of set the stage for this.  When the state of Maryland took over the Baltimore City Jail was my first real exposure.  I mean as a former police officer, I’d been in and out of jails.  I’d been in and out of booking situations.

Spurgeon Kennedy:  Professionally maybe.

Len Sipes:  Professionally, yes.  Thank you for clarifying that.  But I mean spending a lot of time within a jail, wow, what a chaotic setting.  You’ve got thousands of people being arrested.  They’re moving in the institutions.  They’re moving out of the institution.  It’s chaotic.  It’s dangerous.  You’ve got people who are high on some sort of substance.  You’re processing them.  You’re booking them.  You’re making decisions in terms of who to keep and who not to keep.  It’s a very chaotic, loud, noisy situation, and I sat there as people made decisions based upon instruments as to who they’re going to let go on bail.  Who they’re going to let go on pretrial supervision, who they’re going to let go on their GPS or home monitoring, and, you know, we’re not talking about a business offices where it’s nice and quiet and sedate.  We’re talking about a very loud noisy chaotic place, and in that very loud noisy and chaotic place, we’re making decisions that could have an impact not just on public safety, but as to whether or not that person returns for trial.

Spurgeon Kennedy:  Uh-huh.  That’s a wonderful point that you bring up because when we put this document together, and now the folks who helped us draft it, or the people who actually did draft it are the Pretrial Directors Network of NIC.  One of the questions that came us is can you measure performance if you don’t have the things in place to make that performance happen?

Len Sipes:  Right.

Spurgeon Kennedy:  For example, if you are screening defendants for release consideration, if you aren’t using a validated risk assessment…  Something that takes into account factors that have been shown my research to be related.

Len Sipes:  Uh-huh.

Spurgeon Kennedy:  The failure to appear and re-arrest.  Can you really make this measure?  And so part of what we’re trying to do with the measuring what matters document is to say to pretrial programs out there, you have to adopt good business practices.  It’s not just putting a number and a target out there.  It’s also sayings, we need to have in place the things that will make us meet these targets and do a job, and so that validated risk scale that helps you identify who the good risks are, who the bad defendants who need to be incarcerated pretrial are.  You have to have those things in place so that that was something that was very much a part of this document.

Len Sipes:  Lori, we’re going to go over to you for a second in terms of talking about that risk assessment instrument, but amazingly, we’re halfway through the program.  These programs go by awfully quickly.  Spurgeon Kennedy, director of research, analysis and development for the Pretrial Services Agency for the District of Columbia, is our guest, WWW.DCPSA.GOV, an 88 percent return rate here in the District of Columbia, one of the highest in the United States.  Lori Eville, correctional programs specialist for the National Institute of Corrections of the US Department of Justice is our other guest, WWW.NICIC.GOV.  Okay Lori, so we’re going to go over to you in terms of talking about instruments.  I’ve been in the system for 42 years.  This whole concept of an objective series of measurement has always confused me.  It’s not my background.  It’s not my forte, but it’s bottom line seems to be from the criminological community, from NIC, is that we really can, through objective instruments, objective questions, figure out who’s a danger to society.  Who’s not.  Who’s a danger as to whether or not that person is going to return for trial?  Who’s not going to return for trial.  These instruments give us a lot of information about that individual, correct?

Lori Eville:  That’s correct, and it gives us objective information that has a body of research or tested, so that we can make statistical probabilities of whether this person will appear in court or they’re a public safety threat, if you will.  You know, I liked your, what you said about jail being this chaotic place in which you’re having to make these very difficult decisions around release, and I think one thing that objective validated risk, assessment tools do for people working in jails and pretrial programs is that they serve to provide some objectiveness to sort of coral this chaos so that people are not left to think own subjective thoughts around a person’s risk.

Len Sipes:  Right, right.

Lori Eville:  And one of the things that often I think are misunderstood around validated assessments is that we’ve given all of our discretion to this piece of paper.  That this piece of paper, these measures are telling us who’s safe and who’s not, and I think that what we need to understand is that it is a tool within the overall tool box to help make good sound release decisions that are in the public’s interest, and again to make the connection to the document.  If we then have local measures and systems set up, we can then see if in fact we are making good decisions on release.

Len Sipes:  Right.  Just a point of clarification, either one of you can come in on this.  We create these instruments and we do this analysis in this noisy chaotic overcrowded place, and then that information is presented to a Commissioner, an Officer of the Court, and he or she uses that recommendation based upon that instrument to decide whether or not that person is released or kept.  If the person is kept, generally speaking, in most states, you have twenty-four hours to present yourself before a judge, and the judge reviews the evidence, reviews the case, and then makes another decision as to whether or not you’re kept or released or the conditions of the release.  Do I have that correct?

Spurgeon Kennedy:  Yes.

Len Sipes:  Okay.  So it’s not just us within the criminological community or us as correctional professionals, so it’s not us making the decisions.  Basically what we’re doing is preparing the paperwork for the courts.

Spurgeon Kennedy:  Right.  We’re helping the court make a more informed decision about release and detention.

Len Sipes:  And in essence, to go back to what we did before measuring what matters in the National Institute of Corrections, WWW.NICIC.GOV.  You can get the document there, is to basically guess.  Is that fair or unfair?

Spurgeon Kennedy:  There has always been criteria that courts have used.  There have always been criteria that pretrial programs have used.  Up until recently, I think you can make the argument, and I would, that those criteria, may not have been related to failure to appear and re-arrest as much as we tended to believe.

Len Sipes:  But unless, Lori would say, that unless it’s a validated risk assessment instrument, a validated instrument uses the board, it’s still guessing.  Now guessing may be an unfair term, but in essence, you know, without a validated instrument to guide us, it really is a matter of presumption, again.

Spurgeon Kennedy:  Well like to call that clinical assessments, but that’s still guessing.

Lori Eville:  Correct, yes.

Spurgeon Kennedy:  And with a we’ve found and there’s about 60 years of research on this, is that a good validated risk assessment beats clinical judgment about these decisions every time.

Len Sipes:  Right.

Spurgeon Kennedy:  It’s beyond a debate now.  One of the things that I’ll put out there, and there’s an organization that would really be helpful for your listeners to know is the Pretrial Justice Institute.  They have put out a couple of publications on the state of science in pretrial programming.  One of them looks at risk assessments and recommendations, and it’s really a summary of risk assessment of research that has been done over the last ten, perhaps 15 years, and it shows you the common factors that have been coming out of these risk evaluations and shows you that the risk levels that have been coming out, and the factors that predict failure really are becoming common among a lot of jurisdictions, and it’s a good read for people who are really interested in what we’re finding out about risk.

Len Sipes:  Well, let’s talk about those because the audience is going to want to know, well what are the findings?  First of all there are a certain category of people who in all probability are going to be kept.

Spurgeon Kennedy:  Yes.

Len Sipes:  You commit murder.  You commit rape.  You commit domestic violence.  You’re going to stay in all probability.  You’re not going to be released in pretrial.  You’re going to be held in the jail setting.  You’re going to be held in the jail setting, by the way, let me clarify jails.  Jails are also places to hold people on a pretrial basis.  They also are places where people serve short sentences.  Prison is where they ordinarily serve sentences of a year or more, so within that jail, it’s just not pretrial people, it’s people serving short sentences so that limits the amount of beds that you have.  I wanted to make that clarification.

Spurgeon Kennedy:  You also have people, excuse me.

Len Sipes:  No, please.

Spurgeon Kennedy:  Who are also waiting transfer to prisons.

Len Sipes: To prisons.

Spurgeon Kennedy:  In jails and that’s becoming a much larger population as state prisons becoming over crowded.

Len Sipes:  Well, thank you for bringing that up because in a lot of states, it’s a very serious issue.  I mean, 20 percent, 25 percent, 30 percent of your population could be people waiting transfers to state prisons because state prisons are crowded.  Okay, so having said that, what are the other factors?  Okay, so they’re predictive and after years of looking at this, we know that they’re predictive.  What predicts a person returning for trial and not posing a risk to public safety, and I would imagine if he or she has family in the community, owns a home in the community, has a job in the community, that person, more than likely, is going to return for trial.

Spurgeon Kennedy:  Well, community information is not as correlated to failure as we believed.

Len Sipes:  Really?

Lori Eville:  In the early beginnings, that’s correct.

Spurgeon Kennedy:  In the early beginnings of pretrial programs, in the early 1960s, the risk classifications tools that were used relied very heavily on what were very middle class values.

Len Sipes:  Yeah.

Spurgeon Kennedy:  If you looked like me, and if I was middle class, and you wouldn’t be a risk to be released.

Len Sipes:  Right.

Spurgeon Kennedy:  What we’re finding is that a lot of the factors associated with risk are things such as prior failures to appear, substance abuse usage, mental health issues.

Len Sipes:  Okay, if you haven’t appeared, if you skipped your trial in the past, that indicates that you’re going to do it again.  If you’re on drugs when you were arrested and have a drug history.  That’s then a greater chance of not complying.  I think I saw a statistic from Pretrial Services Agency for the District of Columbia that there’s a huge difference in the success between those on drugs and those not on drugs.

Spurgeon Kennedy:  Right.  Those who are nondrug users have much better failure – I’m sorry, appearance rates, and also safety rates.

Len Sipes:  Right.

Spurgeon Kennedy:  than those who are drug users.

Len Sipes:  What are some of the others, either one of you?

Lori Eville:  Well, actually your criminal history is one of them.  Okay.  So if you have convictions in the past, is that on one of them?  The interesting thing you had mentioned about a home.  That is another one that is not necessarily a predictor.

Len Sipes:  That’s interesting.

Lori Eville:  And in fact some of the local validated assessments have actually taken that out.

Len Sipes:  That’s interesting.

Lori Eville:  Because they have not shown that they have a predictive quality to appearing.

Len Sipes:  That’s interesting.  What else has a predictive quality?

Spurgeon Kennedy:  Well those are the big ones.  I think there are about seven or eight that you sees in most of these validated studies, but the question for the local jurisdictions are, A, how do you define these factors.  In Washington DC, for example, a prior failure to appear might be defined a little differently than other jurisdictions depending on how you get data from your court.  The other is whether or not you weight these criteria the same.  A prior failure to appear might be one of the main factors in my jurisdiction.  It might be a lesser factor at other jurisdictions.  So while we see these factors in most risk validation studies, how they’re defined, and how they’re weighted in the final assessment –

Len Sipes:  So one size is not going to fit all in terms of a measurement instrument.  The measurement instrument could be looking at the same variables in Kansas City and San Diego and Washington DC, but yet be interpreted and applied differently.

Spurgeon Kennedy:  Much differently.

Lori Eville:  Based on local culture, differences in jurisdictions, absolutely.

Len Sipes:  That’s interesting.  So everybody – it’s not just one size fits all.  Everybody gets to take a look at this and measure.  But even with that measurement, there’s still a human being that says, I don’t like this outcome.  I think the person should stay.  I think the person is a flight risk, or a public safety risk.  The person can override the instrument, correct?

Lori Eville:  Absolutely.

Spurgeon Kennedy:  One of the performance measures in our measuring with [PH] Matters Peace, in fact, how often the pretrial program actually complies with their own wrist assessment.  We suggest anywhere between a 12 to 15 percent override rate.  If you find yourself in that range of overrides of risk assessment, we think you’re okay.  Because you’re right, risk assessments tools, as great as they are, do not put all defendants in think proper risk places.

Len Sipes:  Well I imagine it’s also going to be concerns regarding race, gender, income, and objective instruments sort of evens everybody out.

Spurgeon Kennedy:  Yes.

Len Sipes:  So if you’re suddenly making exceptions for, you know, 30 percent range for income, you get to ask the question, why?  So the whole idea behind all of this is to generate data and ask questions of yourself as to why you’re getting the results you’re getting, and are people showing up for trial, and are we protecting public safety.  It allows the individual jurisdictions to take a good hard long look at themselves and to improve.

Spurgeon Kennedy:  Exactly.

Lori Eville:  Absolutely.

Len Sipes:  And that’s the bottom line behind what NIC is trying to do across the board in materials of correctional tracking, is to take a good hard long look at yourself and improve.

Lori Eville:  Correct, and you know, this actually – the document, one of the purposes, I mean there are many as we’ve talked about today, but NIC provides a training, as you said, in Colorado for new pretrial directors, and we’ve found that we had new pretrial directors that were being put into these departments and didn’t know really what should they be measuring.  What were their guiding principles, and so it’s part of what NIC is also doing to help bring on new directors and shaping really how pretrial is functioning across the United States.

Len Sipes:  Well, we’re just about at the two minute warning level.  Any final thoughts?  Any quick final thoughts?  Spurgeon?

Spurgeon Kennedy:  Well, only that I really encourage not only pretrial programs, but also any policy maker, or any criminal justice practitioner to take a look at this document.  It should show you how your pretrial programming should work in your local jurisdictions.  It should help you determine what your goals and missions are and how to make sure that you’re doing the best job that you can in order to ensure public safety, and to help the court operate as efficiently as it can.

Len Sipes:  Lori, 15 seconds.

Lori Eville:  I would say go find the document, at the NIC website.  Contact me.  NIC is here to answer any of your questions or help assist your jurisdiction in getting the outcomes that they want.

Len Sipes:  Thank you to the both of you.  Spurgeon Kennedy, director of research, analysis and development at the Pretrial Services Agency for the District of Columbia.  WWW.DCPSA.GOV.  Lori Eville, correctional programs specialist for the National Institute of Corrections for the US Department of Justice, WWW.NICIC.GOV, WWW.NICIC.GOV.  The document we’ve been talking about today is called Measuring What Matters:  Outcome and Performance Measures for the Pretrial Services Field.  Ladies and gentlemen, this is DC Public Safety.  We appreciate all of the comments that you give us.  The letters, the phone calls, the emails.  Please contact us, feel free to contact us with suggestions, programs, program suggestions, criticisms, comments, and if you please, everybody, have yourselves a very, very, pleasant day.

[Audio Ends]

%d bloggers like this: