Offender Rehabilitation: From Research to Practice 1997-01

Offender Rehabilitation: From Research to Practice 1997-01 PDF Version (222 KB)
Table of contents

By James Bonta, Ph.D.

The views expressed are those of the author and are not necessarily those of the Ministry of the Solicitor General of Canada.

Public Works and Government Services Canada
Cat. No. JS4-1/1997-1
ISBN: 0-662-62937-X

Executive Summary

 More than twenty years ago there was a general disillusionment about the effectiveness of offender treatment programs. Various social and political events in the 1970's cast doubt in the public's mind about the ability of governments to control crime and promote justice. As a consequence, there was a shift toward the increased use of sanctions as the means for crime control. However, consideration of the evidence indicates that offender treatment programs can "work" and that sanctions have a relatively small impact on the recidivism of offenders.

In recent years there have been considerable advances in our knowledge about the characteristics of effective treatment programs. Two important principles of effective treatment are the Risk Principle and the Need Principle. The Risk Principle states that the level of treatment should match the risk level of the offender. That is, higher risk offenders require intensive levels of treatment services while low risk offenders require minimal levels of treatment. The Need Principle identifies two type of offender needs: 1) criminogenic and 2) noncriminogenic. Criminogenic needs are offender risk factors that when changed are associated with changes in recidivism. Effective offender treatment programs are those that target criminogenic needs.

The research on offender rehabilitation is translatable into practice. The first step in effective programming is to use well validated measures of offender risk and criminogenic needs. There are today, a number of objective risk-needs scales that can be used for this purpose. These instruments provide a reliable means for differentiating high risk offenders from low risk offenders (Risk Principle) and assessing dynamic risk factors (Need Principle). Correctional jurisdictions are encouraged to develop and use these instruments.

When treatment programs are compared with criminal justice sanctions, the findings show treatment more likely to reduce recidivism. Even detailed analyses of types of sanctions (e.g., length of sentence, boot camps, etc.) show no one particular sanction as significantly effective in reducing recidivism. The evidence is persuasive. If we are to enhance community safety, offender rehabilitation programs that follow the principles of effective treatment are most likely to meet with success.

Offender Rehabilitation: From Research to Practice

This paper attempts to translate into practice what we know from the research on effective interventions with offenders. During the past twenty years, significant progress has been made in our understanding of the factors which influence criminal behaviour. We now know that treatment can be successful in reducing criminal behaviour and we have better knowledge as to what are the effective treatment ingredients.

Although our intellectual understanding of criminal behaviour has progressed significantly from only a couple of decades ago, it is often difficult to implement the research findings in a way that is easily understood and readily accepted by the field. Too often social science progress is seen as an abstract enterprise that has little relevance to "real life" and that recommended solutions are too costly to implement. I will try to show that much of what researchers have learned can be put into daily practice and in a cost-effective manner.

There are two sections to this paper. First, a brief overview of the research on the effectiveness of offender rehabilitation is given and contrasted to the research on the effectiveness of criminal justice sanctions. Next, I take the research results and develop some general ways of implementing the findings into practice.

Research on Effective Offender Rehabilitation

One of the prevailing myths in corrections is that offender rehabilitation does not "work" and that it has never been effective in reducing recidivism. This myth was greatly enhanced by Lipton, Martinson and Wilks' (1975) review of the offender rehabilitation literature and their conclusion that treatment is ineffective. Subsequently, many critics of offender treatment programs made selective references from previous reviews of the rehabilitation literature (e.g., Bailey, 1966; Kirby, 1954) charging that providing treatment to offenders never did demonstrate reductions in recidivism.

A close examination of the literature reviews that supposedly did not support the efficacy of offender treatment indicates that some treatments do work (Andrews & Bonta, 1994). Beginning with the first review article (Kirby, 1954) and continuing to Logan's (1972) review, the majority of studies showed reductions in offender recidivism (see Table 1). However, the 1970s and the decades following were not a fashionable time for ideas of rehabilitation. Nevertheless, research on offender treatment programs continued to show that some treatments reduced recidivism and that there was a growing understanding of the conditions necessary for effective interventions.

Table 1: Reviews of the Effectiveness of Rehabilitation
Review n % effective
Kirby (1954) 4 75
Bailey (1966) 22 60
Lipton et al. (1975) 231 40-60
Logan (1972)* 18 50

* Logan reviewed 100 studies, but only 18 were studies of counselling/therapy with a comparison group. Note: n = number of studies reviewed.

The Key Characteristics of Effective Intervention

In 1990, Andrews, Bonta and Hoge described the common characteristics of offender programs that reduced recidivism. They listed four important principles.

The first principle was called the Risk Principle. An effective treatment program must be able to differentiate offenders in their risk to re-offend and then match their risk to level of service. Higher risk offenders require more intensive services while the lower risk offenders require very little or no services. There is evidence to suggest that intensive levels of services with low risk offenders either has no effect on recidivism or, may even increase recidivism. Thus, reliable risk assessment is important not only for monitoring and release decisions but also for the delivery of effective treatment.

Simply matching levels of service to offender risk level is insufficient for effective programming. The service must address the needs of offenders. The second principle, the Need Principle, makes the point that there are two types of offender needs: criminogenic and noncriminogenic. Criminogenic needs are the offender needs that when changed, are associated with changes in recidivism. For example, substance abuse and employment problems are criminogenic needs. They may serve as treatment goals which, if successfully addressed, may reduce recidivism. Anxiety and self-esteem are examples of noncriminogenic needs. Decreasing anxiety or increasing self-esteem is unlikely to impact future criminal behaviour.

Needs of Offenders
Criminogenic Non-Criminogenic
Procriminal Attitudes Self-Esteem
Criminal Associates Anxiety
Substance Abuse Feelings of Alienation
Antisocial Personality Psychological Discomfort
Problem-Solving Skills Group Cohesion
Hostility-Anger Neighborhood Improvement

The third principle is the Responsivity Principle. There are certain personality and cognitive-behavioural characteristics of the offender that influence how responsive he/she is to types of treatment and how that treatment is delivered. In general, cognitive-behavioural treatments are more effective than other forms of treatment (e.g., psychodynamic, client-centred). But, a cognitive-behavioural treatment program, in and of itself, may not reduce offender recidivism. If the program fails to target criminogenic needs (Need Principle) and with the appropriate intensity (Risk Principle), there may be little effect.

Professional Discretion is the fourth principle. Some offenders may present unique characteristics and situations that are not adequately considered by the other three principles. For example, some sex offenders score low risk on many objective risk instruments but other factors known to the professional may suggest otherwise. As an example, a child molester who is in a position of caring for children may present a special risk not considered by a general offender assessment instrument.

Recently, Andrews has added a fifth principle, Program Integrity. Conducting the treatment in a structured manner, according to the principles outlined and with enthusiastic and dedicated staff are the features of this principle.

Treatment Principles

Evaluating the Treatment Principles

The principles of effective treatment were applied in a meta-analytic literature review. In a meta-analysis, the results of studies are transformed into a common metric which allows the findings to be compared in a quantitative manner. Andrews, Zinger, et al. (1990) argued that not all treatment programs are equivalent. In fact, the words "treatment" and "rehabilitation" are used very loosely in the field. For example, a boot camp may be called a "treatment" just as a cognitive-behavioural anger management program is called treatment. Therefore, it was important to separate criminal justice sanctions from programs that deliver a direct service to the offender.

Furthermore, not all direct service treatment programs are on an equal footing. Andrews, Zinger et al. (1990) argued that treatment programs could be categorised into two sets: a) appropriate, and b) inappropriate. Appropriate programs are those that adhere to the principles of effective treatment. Inappropriate programs violate one or more principles. For example, a treatment program that targeted the self-esteem of low risk offenders with considerable intensity using psychodynamic therapy would violate the first three principles.

Types of Correctional Treatments
Type Description
Sanction Judicial disposition, e.g., restitution, shock probation
Inappropriate Examples:
  1. unstructured individual/group
  2. scared straight
  3. nondirective vocational
  4. inattention to risk/needs
Appropriate Attention to risk, needs, responsivity

Andrews, Zinger, et al. (1990) reviewed 154 treatments that included a comparison group and reported post-program recidivism. Thirty were criminal sanctions (e.g., probation vs prison), 54 were appropriate treatments, 38 inappropriate and the remainder could not be categorised. Analysis of the data showed that only appropriate treatments, those following the principles previously outlined, were associated with reduced recidivism. On average, recidivism decreased by 50%. Criminal sanctions and inappropriate treatments actually showed small increases in recidivism.

Intervention Type and Recidivism

Recidivism by Intervention Type (sanctions +6, inappropriate +5, appropriate -30)

The Andrews, Zinger, et al. (1990) finding that treatment can and does reduce recidivism is not an isolated finding. Lösel (1995) summarised 13 meta-analytic reviews and concluded "all meta-analyses confirm a positive overall effect" (p. 102). Ignoring the offender rehabilitation literature is no longer empirically justifiable.

Criminal Sanctions and Offender Rehabilitation

Almost all countries have formalised rules for applying sanctions to those who violate the rules. These sanctions are intended to reflect society's abhorrence of crime and to address notions of justice by fitting the punishment to the crime. In many Western nations, there is also the hope that sanctions will function as deterrents and reduce future crime by the offender.

The meta-analysis of offender "treatments" by Andrews, Zinger, et al. (1990) found that criminal justice sanctions are not very effective deterrents. Instead, sanctions were associated with more crime! Some may argue that Andrews and his colleagues repeated the same mistake made by Martinson and others who lumped all treatments into one category. In this case however, all sanctions were treated the same. Perhaps, if criminal sanctions were differentiated then we would find some sanctions to "work" better than others in reducing recidivism.

There have been two reviews of the literature on criminal justice sanctions and its impact on recidivism. Cullen, Wright and Applegate (1996) conducted a narrative literature review of community sanctions including intensive supervision programs, electronic monitoring and boot camps. They concluded:

Intermediate punishments are unlikely to deter criminal behavior more effectively than regular probation or prison placements (p. 114)

Gendreau and Goggin (1996) completed a meta-analysis of the literature and were led to similar conclusions. Their findings are summarised in Table 2. None of the sanctions investigated were associated with reduced recidivism of the magnitude found in appropriate treatment studies. A few sanctions were even associated with increased recidivism (fines, restitution). Considering the reviews by Cullen et al. (1996) and Gendreau and Goggin (1996) we have very little reason to believe that any type of criminal sanction would have a significant deterrent effect. Rather, there is a growing recognition that sanctions may reduce recidivism, but only when a treatment component is added (Andrews & Bonta, 1994; Gendreau et al., 1994; MacKenzie et al., 1995).

Table 2: The Effectiveness of Criminal Justice Sanctions
Sanction Studies r
Fines 5 -.07
Shock incarceration 46 -.02
Scared straight 15 -.02
Intensive probation 38 -.01
Drug testing 7 .02
Electronic monitoring 8 .02
Restitution 19 .06
Total 138 .00

(from Gendreau & Goggin, 1996)

Implementing the Results of Research

Summarising the key findings from the research on offender rehabilitation, the following can be said:

  1. Direct treatment services are more likely to reduce recidivism than criminal sanctions.
  2. Effective treatment programs follow the Principles of Risk and Need.
  3. Effective treatment programs are cognitive-behavioural in nature.

Implementing these ideas require a number of steps and the first, and often neglected step, is to make an organizational commitment to the value of offender rehabilitation. Many agencies and correctional managers verbalize their desire to promote rehabilitation but often their actions fail to support their words. Staff are told to "help" their clients, but are not given the time or resources to work with their clients. Clinical professionals are hired, but under-utilised. Treatment programs exist on paper but not in practise. At the other extreme, some jurisdictions in the United States clearly do not hold rehabilitation as a valued activity. Here, probation has become almost like a quasi-police force where staff carry weapons and probationers are simply monitored to make sure that they are complying with court-ordered conditions.

We cannot help but underscore the importance of this first step. Before a probation agency can deliver high quality treatment services, it is essential that the organization accepts the value of rehabilitation, communicates this value to staff, and provides the support for delivering the services. This is not an insurmountable step. An example of making the commitment and communicating to staff the importance of offender rehabilitation can be seen in the Mission Statement of the Correctional Service of Canada. One of the five Core Values described in the Mission statement recognizes the offender's potential to live as a law-abiding citizen and to achieve this goal requires programs which address the offender's needs.

Research shows that effective rehabilitation attends to the principles of Risk and Need. The Risk Principle states that the level of service must match the risk level of the offender. In order to do this properly, the correctional agency must be able to reliably differentiate offender risk levels. It must be able to identify the higher risk offenders who require more intensive services. Although this sounds easy, few jurisdictions conduct state-of-the-art offender risk assessments.

Bonta (1996) described three approaches used in offender risk assessment. The first, which he called first generation risk assessment, is essentially subjective in nature. The professional, based upon his/her training, makes a judgement as to the offender's risk to re-offend. The criteria for judgement and the process by which it is made, is not publicly observable. Subjective assessments often vary from staff to staff and are limited only by broadly worded policy and procedural statements. Professional judgement almost always performs poorly in prediction studies. Unfortunately, this approach to offender risk assessment is widely used.

Second generation risk assessment involves objective criteria that have been empirically linked to recidivism. They are generally referred to as actuarial risk scales. The typical risk scale may have 5 to 10, mostly criminal history items. The second generation risk assessments perform better than subjective approaches. At the very minimum, any correctional agency that has reducing recidivism as one of its goals, must use a second generation risk scale. The scale would provide a reliable means of differentiating the higher risk offender who requires treatment from the lower risk offender who needs only minimal levels of service.

The problem with second generation risk assessments is that because they are mostly comprised of static, criminal history items, they provide little information on criminogenic needs. The Need Principle calls attention to criminogenic needs which should be targeted by treatment services. For a correctional organization to deliver effective services, it must have a way of assessing criminogenic needs. The solution can be found in the use of third generation assessments.

Third generation assessment instruments provide objective, empirically validated assessments of criminogenic needs. To our knowledge, there are only three instruments in operation today which approximate the assessment of criminogenic needs. They are the Wisconsin classification system (Baird, Heinz & Bemus, 1979), the Community Risk-Needs scale of the Correctional Service of Canada (Motiuk & Porporino, 1989), and the Level of Service Inventory - Revised (Andrews & Bonta, 1995). These scale not only include criminal history items but also measure such offender needs as substance abuse, employment and accommodations.

Approaches to Offender Risk Assessment

First Generation

Second Generation

Third Generation

The third important finding from the research is that cognitive-behavioural interventions enhance the effectiveness of treatment programs. Cognitive-behavioural treatment programs have the following characteristics: 

  1. The goal is to train behavioural skills.
  2. The programs are clearly structured.
  3. The therapist is interpersonally warm, socially skilled but firm and consistent.
  4. The therapist models the appropriate behaviour.
  5. The therapist provides feedback. Prosocial behaviour is reinforced and antisocial behaviour is discouraged.

The effectiveness of cognitive-behavioural interventions can partly be explained by the Responsivity principle (Bonta, 1995). Service providers must attend to the personality and cognitive characteristics of their clients in order to maximize the benefits of treatment. Since many offenders tend to be concrete oriented in their thinking, not very verbal, and inadept with certain prosocial skills (e.g., work, interpersonal), traditional counselling approaches (e.g., client-centred and psychodynamic therapies) have been ineffective with offenders. These traditional therapies require clients to be verbally skilled and already have prosocial behaviours in their repertoire.

In summary, in order to reduce offender recidivism, we make the following recommendations:

  1. commit to the value of offender rehabilitation and give careful attention to the research;
  2. use an objective and empirically validated risk-needs instrument to match risk to the level of service;
  3. deliver cognitive-behavioural interventions that target criminogenic needs

These recommendations are feasible but, they are also not easy to realize. Despite the difficulties, efforts to design and implement programs based on the research will likely produce direct benefits to the client and the community in the form of reduced recidivism.


Andrews, D. A., & Bonta, J. (1994). The psychology of criminal conduct. Cincinnati: Anderson.

Andrews, D. A., & Bonta, J. (1995). The Level of Service Inventory-Revised. Toronto: Multi-Health Systems.

Andrews, D. A., Bonta, J., & Hoge, R. D. (1990). Classification for effective rehabilitation: Rediscovering psychology.  Criminal Justice and Behavior, 17 , 19-52.

Andrews, D. A., Zinger, I., Hoge, R. D., Bonta, J., Gendreau, P., & Cullen, F. T. (1990). Does correctional treatment work?  A clinically-relevant and psychologically informed meta- analysis. Criminology, 28, 369-404.

Bailey, W. C. (1966). Correctional Outcome: An evaluation of 100 reports. Journal of Criminal Law, Criminology and Police Science, 57, 153-160.

Baird, S. C., Heinz, R. C., & Bemus, B. J. (1979). Project Report #14: A two year follow-up. Madison, WI: Department of Health and Social Service, Case Classification/Staff Deployment Project, Bureau of Community Corrections.

Bonta, J. (1995). The Responsivity Principle and offender rehabilitation. Forum on Corrections Research, 7(3), 34-37.

Bonta, J. (1996). Risk-needs assessment and treatment. In A. T. Harland (Ed.), Choosing correctional options that work (pp. 18-32). Thousand Oaks, California: Sage. 

Cullen, F. T., Wright, J. P., Applegate, B. K. (1996). Control in the community: The limits of reform? In A. T. Harland (Ed.), Choosing correctional options that work (pp. 69-116). Thousand Oaks, California: Sage.

Gendreau, P., Cullen, F. T., & Bonta, J. (1994). Intensive rehabilitation services: The next generation in community corrections. Federal Probation, 58, 72-78.

Gendreau, P., & Goggin, C. (1996). Principles of effective programming with offenders. Forum on Corrections Research, 8(3), 38-40.

Kirby, B. C. (1954). Measuring the effects of treatment on criminals and delinquents. Sociology and Social Research, 38 , 368-374.

Lipton, D., Martinson, R., & Wilks, J. (1975). The effectiveness of correctional treatment: A survey of treatment evaluation studies. New York: Praeger.

Logan, C. H. (1972). Evaluation research in crime and delinquency: A reappraisal. Journal of Criminal Law, Criminology and Police Science, 63, 378-387.

Lösel, F. (1995). The efficacy of correctional treatment: A review and synthesis of meta-evaluations. In J. McGuire (Ed.), What works: Reducing reoffending: Guidelines from research and practice. Chichester, England: Wiley.

MacKenzie, D. L., Brame, R., McDowall, D., & Souryal, C. (1995). Boot camp prisons and recidivism in eight states. Criminology, 33, 327-357.

Motiuk, L. L., & Porporino, F. J. (1989). Field test of the Community Risk/Needs Management scale: A study of offenders on caseload. R-06. Ottawa: Correctional Service Canada.

Date modified: