The challenge of change: Correctional programs and evidence-based practices
Latessa, Edward J. (2004). The Challenge of Change: Correctional Programs and Evidence-Based Practices. Criminology & Public Policy, Volume 3 Number 4, pp 547-560.
Latessa examines the correctional programs and treatment efforts across the US. His observations on the challenge of change in corrections suggest that having a strong leadership is a necessary ingredient for the change process to occur and be sustained over time.
Over the past ten years or so, much of my effort has been devoted to working with correctional agencies around the country. During this time, I have conducted hundreds of workshops and training sessions on evidence-based practices and programs for offenders. My current and former doctoral students and I have assessed over 360 correctional programs of all shapes and sizes, including those serving adults as well as juveniles, those in prisons, and those conducted in the community. During this time, I have seen some of the worst that corrections has to offer, as well as some of the best. I have seen programs that I would not refer anyone to, and programs that have had a demonstrated effect on offender behavior and recidivism. I have also had the opportunity to work with many dedicated and committed people, and I am indebted to those who have not given up the quest for new knowledge and advancement in correctional research. Among those whom I have had the pleasure to work with and learn from include some of the giants of correctional rehabilitation: Paul Gendreau, Don Andrews, and Frank Cullen. There are many others, but space does not permit me to thank them all individually. I am forever in their debt. This brief prelude brings me to the purpose of this essay: to discuss the challenge of change in corrections.
Let me also say that my comments are going to be narrowly limited to a discussion of correctional programs and treatment efforts, which is not to say that the debate over the use of incarceration, sentencing practices, and other policy-related issues is not closely linked to rehabilitative efforts. They obviously are; its just that my experience and research has been more focused on studying and, hopefully, improving correctional programs, rather than on more policy-related concerns. Its not that I have no interest in changing systems; its just that I have set my sights much lower, and I am blissfully content if I can improve a small program. It is within this context that I want to present some observations I can make from venturing outside the ivory tower into the field of corrections. I will address several specific topics. First, I begin by discussing some lessons for correctional change; that change is difficult and that there is political context that needs to be addressed. In this section, I also discuss organizational readiness, the importance of leadership, and the challenge of overcoming the armchair quarterbacks that pervade the field. In the second section, I will speak to some of the reasons that researchers and academics share the blame and offer some suggestions for remedying this problem. Finally, I will end by offering some examples of states that have been trying to use evidence-basked knowledge to improve correctional programs.
LESSONS FOR CORRECTIONAL CHANGE
CHANGE IS HARD EVEN FOR US
First, we have to accept the premise that change is difficult. I often tell correctional staff who complain about the difficulty of changing offender behavior to try to change something about themselves. For those who have tried to lose weight, quit smoking, eat fewer sweets, or exercise more, you know it is not easy, and heaven knows many of us try. Why would we think that it would be easy for a correctional system or an agency to change, especially given the relative comfort that exists in maintaining the status quo? It is important to remember that corrections often operates under the modus operandi of “if nothing bad happened yesterday, do the same thing today.” As with people, I suspect that organizations often view change reluctantly, and ultimately ask, “why should we change?” Researchers, of course, like to point to the data and evidence as the logical reason to change; however, when you consider that 80% is the median rate for nonadherence with advised health care practices by health care professionals, why would we expect any different behavior from correctional professionals? Given the general lack of motivation that exists among people it should not be surprising that correctional organizations resist change.
UNDERSTANDING THE POLITICS OF CHANGE
Given this premise, it is important that those with an agenda for change understand the political context that exists at all levels of a correctional organization, be it large or small. Although there are exceptions, first and foremost, politicians and those they appoint, including correctional policy makers, are committed to survival, which usually translates into an aversion to what they perceive as risk taking. It is not that policy makers are necessarily opposed to evidence-based programs and practices, in fact I have found just the opposite to be true, but it is just that they have to be helped to understand the “up-side,” if you will, as to why change can be beneficial. For example, when I ask policy makers to give me an estimate of the percentage of the public that they believe view the primary purpose of prisons to be punishment, they greatly overestimate the punitiveness of the public. As numerous studies have demonstrated, the public’s support for effective programs and rehabilitation is still strong.1 Helping them understand that the public is not monolithically punitive and that the research shows that a large percentage support rehabilitative efforts can be a powerful way to begin breaking down some of the resistance they may have to developing effective correctional programs and alternatives to incarceration. Of course, this is only one aspect of making change politically palatable. As public protection is often seen as a fundamental goal of correctional officials, another important step is to demonstrate to them that using research to improve correctional programs can actually increase public protection, whereas conversely, using approaches that have not been found effective can have the opposite effect. I have found very few policy makers unwilling to at least listen to the empirical research when you frame it within the context of public protection.
Of course, getting policy makers to listen is only the first step. For decision makers, it may involve helping them understand what evidenced-based practices are, that developing more effective alternatives and programs based on research is not necessarily going to put them at odds with public opinion, and that they may in fact be more consistent with what the public wants than they know. Telling them that is not enough, however. It may also mean getting them to understand that effective treatment and incarceration are not always mutually exclusive. The fact is, we are always going to operate prisons and incarcerate a significant number of offenders, but that does not mean we should not be designing effective correctional rehabilitation programs. Similarly, the public will support community alternatives, but they also appear to want something “done” with offenders, and they want those interventions to be effective in reducing recidivism.
If meaningful change is going to occur, it is also important to communicate with correctional staff so that they can begin to understand some of the benefits for them. For example, if you are trying to implement evidenced-based programs in a prison, do you really think that telling the correctional officers that the program may help to reduce recidivism is going to convince them to lend support to your efforts? For these staff, the value of evidence-based programs is the potential reduction of critical incidents and safety issues. For the bean counters, the critical issue may be the cost savings that can accrue. The point is that the change process will require support at all levels of an organization, which will require an understanding of the concerns and issues that confront various staff. This point brings me to my next one, the need for organizational responsivity.
MOVING BEYOND PROVIDING DATA: ORGANIZATIONAL RESPONSIVITY
Why would we think that simply providing someone with facts and data would lead them to change their behavior? It certainly does not work with offenders (or most of us, for that matter), so it should come as no surprise to us that organizations are unlikely to change simply because we give them some information or research findings.
One of the characteristics of effective correctional treatment programs is that they assess an offender’s readiness to change before they begin the actual program. Identifying barriers (and strengths) that a person may have should be part of the assessment process. This information can then be factored into a case plan so that the program can better match the offender to treatment or, as is more often the case, better prepare the offender for the program. The process for an organization should be similar: Assessing an organization’s readiness to change and then developing a strategy or action plan can be an important step in the process. Preparing an organization for change can significantly increase the chances that the changes will be effectively implemented and supported.
LEADERSHIP IS IMPORTANT AT ALL LEVELS
I have conducted workshops for correctional staff, and when I finish, they sometimes say, “you are talking to the wrong group, you need to talk to the policy makers, since they are the ones that make the decisions.” Then when I address the policy makers, I am often told, “our staff need to hear this, since they are the ones who will be responsible for changes.” Of course, this is not a problem for me, because I will talk with whomever will listen, but what I have learned is that if you are going to be successful in moving an organization forward, strong leadership for change needs to be in place at every level. For example, I do not find effective correctional programs in jurisdictions or organizations unless some high-level leader is willing to take charge and make things happen. Without a “hero” or two, it is very difficult for effective programs to emerge, let alone sustain over time. However, although vision and commitment from policy makers is necessary, it is not sufficient. Strong involved leaders and role models are also needed at the program level. Those persons who are responsible for the day-to-day operation of correctional programs need to be directly involved in designing, implementing, and operating their programs. In other words, they need to be “on the shop floor” so to speak. Whether it means cofacilitating groups, carrying a small caseload, or conducting some offender assessments, it is important for program directors to be involved in some aspect of service delivery. I remember one residential facility I assessed a number of years ago. When I asked the program administrator if he was involved in working with offenders, he replied that he always carried a caseload of the five highest risk offenders in the facility. Similarly, I have worked with a large urban probation department that talked for a long time about implementing evidence-based programs. They were having a tough time getting staff to change until the chief probation officer began cofacilitating cognitive behavioral groups and having his supervisors do the same. These leaders not only talked about supporting evidence-based programming, also they practiced it.
These examples illustrate good role modeling and the type of leadership that can have a significant impact on staff and the organizational culture. To further illustrate my point, I would like to use the example of the person who becomes a dean, and the first thing he/she does is quit teaching. The second thing he/she does is tell faculty how important teaching is, and how we have to do a better job. This person is not my idea of a good role model.
EVERYONE IS AN EXPERT
One of the problems with crime is that everyone is an expert. I would dare say that if I studied quantum physics, few people would offer their opinions about how I should go about my business, but because I study criminal behavior and corrections, everyone offers me advice. For example, once I was on a flight and was seated next to an older woman. She asked me what I did for a living, and I made the mistake of telling her I was a criminologist. For the next four hours she told me how to solve the crime problem. Now I just tell folks I am a proctologist and they leave me alone.
Although this story may be humorous, the problem of everyone believing they are an expert about crime affects correctional practice at many levels. From the politician to the case worker, everyone thinks they know how to deal with offenders and what we need to do to “straighten” them out. As an aside, I often ask correctional staff who work with offender’s day in and day out what they think are the major risk factors associated with criminal conduct. They are often all over the map, and needless to say, I am often amazed with the list they come up with. Interestingly, several years ago, my youngest daughter asked me to come to her fourth-grade class and talk about what I did for a living. When I asked the class to tell me why they thought some people got into trouble, they named many of the risk factors supported by the research: anti-social attitudes, hanging around with the wrong friends, personality traits, familial care and supervision, and lack of social support. As the old saying goes, “out of the mouth of babes.” When you combine the “everyone is an expert” problem with the lack of credentialing (both individual and agency), little if any adequate staff training on the skills needed to change offender behavior, and an abundance of ignorance about what the research tells us, is it any wonder that we have so much correctional quackery being practiced?2
WE SHARE SOME OF THE BLAME
As I have gone out into the field, I have also been struck by the failure of researchers and scholars to bridge the gap among theory, research, and practice. I am certainly not the first to voice concern about the relevancy of criminology or to suggest ways to increase it.3 I do, however, believe that research has made a difference, at least in the area of correctional rehabilitation, and I will give some examples later. For now, let me offer some suggestions for how researchers and academics can do a better job of promoting change.
1. Leave the Office. It is almost a given that most practitioners and policy makers do not read the literature or published research (and this certainly helps explain why we have so many correctional programs based on half-baked theories). If we want our research to have an impact on the field, we need to recognize that it may be necessary to leave the office. We have to be willing to attend and present at nonacademic conferences, conduct workshops for local professionals, testify at legislative hearings, and in general be willing to lend our expertise and knowledge when asked to do so. Let me also add that I realize that it is a lot easier to sit in our offices and classrooms and pontificate to our students, but if we expect to have an impact, we need to be willing to get our hands a little dirty. It is not always easy to face a skeptical, if not sometimes hostile, crowd, and although it is unpleasant to be challenged and questioned, it may also be necessary if we are going to win over converts to evidence-based practices.
2. Make research understandable. I have gone to numerous practitioner-oriented conferences where researchers and academics spend all of their allotted time talking about the methodology or statistical techniques they used rather than focusing on their findings and its relevance. Talking about the log ratios and beta weights may be important to other researchers, but it will not do much to transfer knowledge to the field. I find that too often scholars write and talk in such a way as to make their work undecipherable by those who can most benefit from it. If we want to have an impact, it is 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.
3. Include measures of program integrity and quality in our research. Too often, correctional researchers have focused on the offender and their characteristics to explain the results from a study of a correctional program or intervention. Relatively few outcome studies include measures of program integrity or fidelity. We often lump together in outcome studies programs of various quality and implementation efficacy. Not only can the “failures” cancel out the “successes,” but also by failing to measure program characteristics and fidelity we are often unable to explain some of the programmatic reasons why differences in outcome measures may occur. For example, in a recent study of similar types of residential correctional programs in Ohio, we found a wide range of effects based on the quality of the program.4 Although most of the programs were found to have a positive effect on recidivism rates, a handful actually produced increased rates of recidivism. Fortunately, we included a number of measures of program integrity that helped explain these findings. Another example of the importance of program fidelity can be seen in the recent study by the Washington State Institute for Public Policy. These researchers found that several evidenced-based interventions had a significant effect on recidivism, provided they were competently delivered. If not, they actually increased failure rates.5
4. Do a better job of preparing our students. The doctoral students who study with me are well versed in working with practitioners, and I hope that they have learned from it. However, I also recognize that they are not likely to be working in the trenches of corrections. It is the undergraduates that are produced that will occupy most of the positions in the field, and to be blunt, we need to do a better job of preparing them to work in corrections. In my opinion, three important ingredients improve the preparation of undergraduate students to work in corrections: (1) teach them the knowledge base, (2) provide them better skills and competencies, and (3) expose them to other relevant disciplines. Let me touch briefly on these points.
Perhaps I can best illustrate the first point by giving an example. Most of us believe that cigarette smoking is harmful to our health. Why? Well, one might answer, “because of the research that has been conducted.” Yet, I would guess that if only one or two studies were conducted on the health risks of tobacco, we might not be so sure. The reason most of us believe that smoking is harmful to our health is because research has been conducted for decades all over the world by independent researchers that have concluded that if you smoke, it can lead to cancer, heart disease, emphysema, and other health problems. In other words, a body of knowledge exists about the effects of smoking. I would argue that a significant body of knowledge exists in corrections as well. Research on risk factors and correctional treatment has been ongoing for decades, and numerous scholars have reviewed and summarized this research. This is not to say that more research is not needed, or that we have all of the answers (we clearly do not), but we do have a considerable amount of knowledge about criminal conduct and effective (and ineffective) ways to rehabilitate offenders. What we need to do is a better job of teaching our students what we do know and its relevance to the field they have chosen to study, and perhaps someday work in.
The second and third points are related. That is, those students who are going to work in corrections with offenders need to have some core skills, many of which may have to come from related fields, such as psychology, social work, counseling, addictions, and other helping professions. Although much of the responsibility for specific training lies with the agencies that hire our students, I do believe that there are some areas that we can better prepare students, such as giving them a working knowledge about the most effective ways to effect change and shape behavior, an understanding of risk and need factors related to criminal conduct, and an understanding of other competencies that will be important in a correctional setting.
CHANGE IS POSSIBLE: EXAMPLES FROM THE STATES
Now that I have gotten that off my chest, I would like to devote the remainder of this essay to providing some examples of states that are attempting to use correctional research to implement change and improve programs for offenders.
I first visited Oklahoma in 1997. At the time, I was part of a team funded by the National Institute of Corrections that was conducting workshops on evidenced-based practices in corrections. Although it was not uncommon for a state or jurisdiction to seek subsequent technical assistance, Oklahoma’s Department of Corrections wanted to begin to assess the programs it offered offenders, including those operated by the state as well as those under contract with private providers. In 1999, a team of researchers from the University of Cincinnati began reviewing programs throughout the State of Oklahoma. Initially, a total of 29 programs were selected for assessment, including those operated in prisons as well as community based. We used the Correctional Program Assessment Inventory (CPAI) as the evaluative instrument.6 The CPAI assesses program integrity and the degree to which a correctional program meets the principle of effective intervention.7 During our initial review, only 9% of the programs scored “satisfactory.” The remaining 91% scored “needs improvement” or “unsatisfactory.” Although some states might have sent us packing at this point, Oklahoma officials made a decision to not only continue the process, but also to institutionalize it as part of their efforts to improve the programs and services they offered offenders. Those programs that were not satisfactory were required to develop action plans on how they were going to address deficiencies and improve their programming. Programs were given specified time periods to correct deficiencies, and reassessments were scheduled. Subsequent program assessments indicated that Oklahoma was able to dramatically improve the quality of its correctional programs, at least as measured through the CPAI. The most recent results indicate that 79% of the programs in Oklahoma are now rated as “satisfactory” or higher, and none of the programs are “unsatisfactory.”
Change certainly was not always easy for many of these programs, but there are several reasons it occurred. First, strong committed leadership was in Oklahoma. Second, specific and clear direction was provided, based on research, and third, training and technical assistance was provided programs throughout the process. Granted, Oklahoma still has a very high incarceration rate, and its sentences are among the longest in the country, but at least it now offers offenders correctional programs that are based on evidence, and the principles of effective intervention.
The second example I would like to highlight is the State of Oregon. Over the past few years, the Oregon Department of Corrections under the leadership of Dr. Ben deHaan sought to develop and implement research-driven practices and programs. In the summer of 2003, I was asked to testify to a joint Oregon Judiciary Committee hearing on the use of evidenced-based programs for offenders. The Oregon legislature subsequently passed SB 267, which requires prevention, treatment, or intervention programs that are intended to reduce future criminal behavior in adults and juveniles or to reduce the need for emergency mental health services to be evidence-based. Furthermore, by 2005, 25% of funds spent by the Oregon Department of Corrections, Youth Authority, the Department of Human Services, the Criminal Justice Commission, and the Commission on Children and Families have to be allocated to evidenced-based programs. By 2007, the amount increases to 50%, and by 2009, 75%. While this state is the first state I know of to statutorily require evidenced-based programs for offenders. I suspect it will not be the last, especially as states continue to wrestle with budget deficits.
In recent years, Ohio, like many states, has been experiencing significant budget shortfalls, and most state agencies, including corrections, have been hard hit. Ohio has also made a significant investment in residential programming for offenders, spending over $89 million this fiscal year for halfway houses and community-based correctional facilities.8 As the budget for these programs has grown, so has the demand by the legislature to justify these expenditures by determining the effectiveness of these programs in reducing recidivism. As fate may have it, I happen to be in the research business, and I received the contract to conduct a study of all of the residential correctional programs funded by the state. In the summer of 2002, we completed the largest study ever conducted of residential correctional programs. Over 13,000 offenders were included in our study of 38 halfway houses and 15 CBCFs. Results from the study showed that treatment effects were strongest for higher risk offenders, and that for all but a handful of programs, the recidivism rates for low-risk offenders actually increased as a result of the programming (Lowenkamp and Latessa, 2002). Although the findings from our study mirrored what one would expect given the “what works” research that exists, the point of my example is to show how research can be used to change correctional practice. As a result of this study, Ohio has enacted a number of policy changes. These include:
* All programs must administer an assessment tool within five days of intake to measure risk level, determine case planning strategies, and identify special needs, such as mental health and sex offender.
* All programs need to develop a service delivery model based on individualized risk and needs assessment results. The high-risk offender should receive more intensive and additional services; conversely, the low-risk offender will receive minimal services.
* A cognitive behavioral modality should be adopted, or minimally cognitive programming skills should be implemented within other modalities.
* Criminogenic targets should be addressed in programming.
* Audit standards shall assess both processes and program outcomes. Standards will be based on a performance-based model wherever possible.
* Program evaluations will be conducted every three years.
* Programs shall conduct a Correctional Program Assessment Inventory, or similar instrument, every three years to ensure program fidelity.
Although some critics may see these policies as basic, they constitute major changes in the operation and monitoring of community-based correctional programs in Ohio, all brought about through research.
Many other examples exist of jurisdictions that are paying attention to the research on evidence-based practices in correctional treatment. These examples include Washington, which through the work of Steve Aos and his colleagues at the Washington State Institute for Public Policy have promoted and studied the effects of these efforts. Maine and Illinois were recently awarded demonstration project grants by the National Institute of Corrections to implement and promote evidence-based practices throughout these states. Likewise, the Florida Division of Juvenile Justice has recently begun an extensive effort to promote evidence-based programming throughout the system. Other states that have been paying attention to the research and have been attempting to implement evidence-based programs include Idaho, Iowa, Wisconsin, Indiana, Pennsylvania, Utah, Colorado, and Minnesota. Of course I would be remiss not to mention the Correctional Services of Canada, which has made evidenced-based programming the hallmark of its correctional system. Undoubtedly, other examples exist.
In the above essay, I have shared some observations that I can make about the challenge of change in corrections, at least as it applies to implementing evidenced-based programs. These can be briefly summarized as follows:
* Change is difficult, and many of the reasons that organizations resist change are the same reasons we as people do.
* Giving reasons to change is part of the process, but sometimes it takes some planning and persistence to remove barriers, as well as someone to motivate.
* Having strong leadership from top to bottom is a necessary ingredient for the change process to occur and be sustained over time.
* Everyone may think they are an expert about corrections and criminal behavior, but few are. Rely on the empirical evidence rather than your neighbor.
* Researchers and academics who are interested in having their work used by corrections need to leave the office, make research findings more understandable, and expand the measures we use to evaluate corrections to include program characteristics. We also need to do a better job of teaching our students the body of knowledge in corrections.
* Despite the challenges, change is occurring, and several examples were offered, including Oklahoma, which has improved correctional programs through assessment and evaluation; Oregon, which has legislated evidenced-based programs throughout the correctional system; and Ohio, which has used research to change policy and auditing processes for programs.
I end with a leap of faith. I am not naïve enough to believe that there is a magic wand that can be waived to dramatically change the policies and practices that have lead to over 2 millions people being incarcerated in America, or that we can somehow have a dramatic effect on policies that are embedded in politics, tradition, custom, and imitation. I do, however, believe that we can make a difference, and that change, although difficult, is possible.
1. Cullen has written extensively about public support for correctional rehabilitation. For an excellent summary of this research, see Cullen et al. (2002).
2. For a better explanation of correctional quackery and some of the theories that are used in treating offenders see Latessa et al. (2002).
3. For a discussion of these issues, see Petersilia (2000) and Austin (2003).
4. See Lowenkamp and Latessa (2002).
5. See Barnoski (2004).
6. The CPAI was developed by Paul Gendreau and Don Andrews to assess the integrity of correctional programs. For a discussion of the instrument, see Latessa and Holsinger (1998).
7. Several studies have found significant correlations between scores on the CPAI and recidivism rates. See Holsinger (1999) and Lowenkamp (2004).
8. Figures provided by the Ohio Department of Rehabilitation and Correction.
2003 Why criminology is irrelevant. Criminology and Public Policy 2:557-563.
2004 Outcome Evaluation of Washington State’s Research-Based Programs for Juvenile Offenders. Olympia, Wash.: Washington State Institute for Public Policy.
Cullen, Francis T., Jennifer A. Pealer, Bonnie S. Fisher, Brandon K. Applegate, and Shannon A. Santana
2002 Public support for correctional rehabilitation: Change or consistency. In Julian V. Roberts and Michael Hough (eds.), Changing Attitudes to Punishment: Public Opinion, Crime and Justice. Devon, U.K.: Willan Publishing.
1999 Opening the “Black Box”: Assessing the Relationship Between Program Integrity and Recidivism. Doctoral Dissertation, University of Cincinnati, Ohio.
Latessa, Edward J. and Alexander Holsinger
1998 The importance of evaluating correctional programs: Assessing outcome and quality. Correctional Management Quarterly 2:22-29.
Latessa, Edward J., Francis T. Cullen, and Paul Gendreau
2002 Beyond correctional quackery-professionalism and the possibility of effective treatment. Federal Probation (September):43-49.
2004 Correctional Program Integrity and Treatment Effectiveness: A Multi-site, Program-level Analysis. Doctoral Dissertation, University of Cincinnati, Ohio.
Lowenkamp, Christopher and Edward J. Latessa
2002 Halfway House and CBCF Findings: Program Characteristics. Available online: http://www.uc.edu.proxy.libraries.uc.edu/criminaljustice.
2002 Evaluation of Ohio’s Community Based Correctional Facilities and Halfway House Programs. Available online: http://www.uc.edu.proxy.libraries.uc.edu/criminal justice.
2000 Policy relevance and the future of criminology. In Barry W. Hancock and Paul M. Sharp (eds.), Public Policy, Crime, and Criminal Justice. Upper Saddle River, N.J.: Prentice Hall.
EDWARD J. LATESSA
University of Cincinnati
Edward J. Latessa is Professor of criminal justice at the University of Cincinnati. He is the recipient of the ASC’s 2004 August Vollmer Award for research scholarship that has contributed to justice or to the treatment or prevention of criminal or delinquent behavior.