Correctional Service Canada
Symbol of the Government of Canada

Common menu bar links

Compendium 2000 on Effective Correctional Programming

Warning This Web page has been archived on the Web.

CHAPTER 23

Contributing to Safe Reintegration: Outcome Measurement

LAURENCE L. MOTIUK1


Criminal justice policy makers and practitioners in the field of corrections have a keen interest in reducing the likelihood of repeat offending because of the enormous costs to victims and society. While crime continues to present a serious social problem, changes in legal definitions coupled with reduced public tolerance for serious crimes and focused media attention have led to significant improvements in policing, court processing and corrections.

Being acutely aware that the general public does not fully understand the inner workings of the criminal justice system, correctional service providers are being called upon to provide timely responses and accurate information on the care, custody and safe reintegration of criminal offenders. Realizing too, that the media has stretched public tolerance to the limit for any failure in the community, has meant that correctional service providers are learning everything there is to know about outcome measurement and become actively involved in public relations.

To summarize the problem -- offenders, staff, volunteers, and public opinion exert a significant influence over the realization of correctional service delivery objectives. In particular, the task of safely reintegrating offenders in the community continues to fall squarely on the shoulders of staff and volunteers located in correctional facilities, mental health settings, and the community at large. Certainly, these people are being called upon to deliver more sophisticated services to a clientele constantly changing and for a public that is uncertain. And to top it all off, they must do so in an effective and cost efficient manner as possible. It's a common challenge, but one that's difficult to measure. This chapter tries to address why correctional outcome is so tough to measure and tries to show how we can measure it as best we can.

THE CORRECTIONAL MANDATE

Correctional agencies at the federal, provincial, and territorial levels across Canada share a common aim or purpose. “Contributing to public protection” is well entrenched in the daily activities of these organizations and the minds of staff. A review of missions, mandates and visions of Canadian correctional agencies is a good illustration of how agencies all across the country have responded to national crime control policies (Motiuk, 2001). Essentially, they aid in the safe reintegration of offenders into society by providing education, vocational and personal development opportunities to offenders. Moreover, each of them track and monitor three important areas of correctional outcome -- safety of the public, safety of staff and safety of offenders (see Figure 23.1). Some common indicators used for reporting results in these areas are encircled in Figure 23.1.

Figure 23.1 Reporting Results

.
.
.

 

CARE, CUSTODY, AND SAFE REINTEGRATION

Offender care is concerned with safe healthy environments for those living and working in correctional systems as well as members of the public. Custody refers to the accommodation and management of offenders in correctional facilities that is reasonable, safe, secure and humane and in accordance with the least restrictive option. Reintegration infers that offenders are being safely and effectively returned to the community. While not to discount the importance of care, this chapter is focused primarily on the outcome measurement of the custody and safe reintegration aspects of correctional case management.

Custody

Custody placement allows society to effectively incapacitate dangerous offenders who show little potential for changing behavioural patterns that threaten the safety of the public. However, for less serious offences that normally results in short prison terms, there are few empirically identifiable benefits to recommend the use of incarceration as a penalty. Imprisonment of offenders for minor crimes may provide some compensation to the public and to victims in terms of retribution. At the same time, longer-term issues of public safety are almost completely neglected when incarceration is used as the only form of correctional intervention. An attractive alternative is to examine the possible benefits of redirecting resources currently committed to incarceration for minor offences and/or low-risk offenders to the crime prevention agenda. For example, a greater emphasis could be placed on alternative sentences served in the community with appropriate treatment programming extended to manageable cases. There is now ample research evidence to suggest that well-designed, community-based alternatives to incarceration for many cases offer greater advantages in terms of controlling criminal recidivism.

Some theorists of criminal behaviour point to the impulsive nature of the crimes committed by violent, sex and repeat offenders who simply fail to think before they act. The research literature suggests that impulsivity and risk-taking are important distinguishing characteristics of violent, sex and repeat offenders. Unfortunately, the impulsivity exhibited by these offenders is likely to severely limit the deterrent value of criminal penalties such as incarceration. Ross and Fabiano (1985, p. 162) offer the following:

The research we have reviewed suggest that deterrence may have little reality or meaning for many offenders. Many offenders seldom consider the consequences of their acts. Many underestimate the risk: some are indifferent to the risks; some thrive on them. Many are sublimely optimistic; they believe that they will not be caught; if caught, not convicted; if convicted, not sentenced; if sentenced, not imprisoned; if imprisoned, quickly released. (We hasten to add that such beliefs are by no means unrealistic).

Deterrence, regardless of the type of penalty, may work well for the majority of citizens who possess the cognitive skills necessary to think before they become involved in illegal activities. However, for violent, sex and repeat offenders the threat of incarceration appears to hold little promise for controlling criminal behaviour. Incapacitation resulting from imprisonment only provides a benefit to society while the offender is incarcerated. The incapacitating effect of imprisonment for a given offender ends with their release.

As a program of rehabilitation, incarceration has not shown much success in rehabilitating offenders. Based on an analysis of over 50 studies involving more than 300,000 offenders, Gendreau, Goggin, and Cullen (1999) explored whether prison reduced criminal behaviour or recidivism. The essential conclusion their work purports “no evidence that prison sentences reduced recidivism”. In fact, prison sentences produced slight increases in recidivism and there was some tendency for lower risk offenders to be negatively affected by the prison experience. They conclude, “the primary justification for use of prisons is incapacitation and retribution”. Consequently, without other forms of intervention that directly address criminal behaviour and attempt to instill new patterns of behaviour, custody on its own lacks promise.

In a study exploring the impact of preventative detention on the post-release violent and sexual recidivism of 424 detainees, Motiuk, Belcourt, and Bonta (1995) found that time served on detention did not reduce the likelihood of violent recidivism. This was after controlling for level of risk of reoffending and time at risk in the community. Another analysis by Bonta and Motiuk (1996) of detention cases and court designated Dangerous Offenders found both the courts and correctional agencies tend to equate high-risk violent offenders with sex offenders. Also noteworthy is the absence of empirical work on optimal sentences for violent, sex and repeat offenders. Consequently, more attention should be paid to selection issues for which a considerable body of literature exists on the predictors of criminal recidivism and treatment efficacy.

Recent reviews of accumulated findings from hundreds of published studies on rehabilitation programs for offenders (Andrews, 1995; 1996; Gaes, Flannigan, Motiuk, & Stewart, 1999; Gendreau & Goggin, 1996; McGuire, 1995; Lipsey, 1995; Lösel, 1995; 1996) yield clear empirical evidence of the impotency of criminal sanctions when unaccompanied by appropriate rehabilitative programming. The results of these reviews also suggest that rehabilitation programming that takes place in custodial settings appears to be less effective than programming that occurs in the community. In view of the evidence that better outcomes are reported for those completing treatment and particularly for programs operating in the community, the notion that offenders can be sent to prison to be rehabilitated without treatment and aftercare is questionable.

The ineffectiveness of incarceration alone and the effectiveness of appropriate rehabilitation programming, particularly community-based, in reducing repeat offending continue to be advanced by a growing body of contemporary researchers. Indeed, Tarling (1993) has noted that a change in the order of 25% (of the prison population) would be needed to produce a 1% change in the level of crime. On the other hand, Gendreau and Goggin (1996) have found that prison programs with a great deal of therapeutic integrity can produce recidivism reductions in the range of 20% to 35%. Consequently, criminal justice systems are being challenged to offer more specialized programming and improved case management services to violent, sex and repeat offenders -- a large, diverse and challenging segment of the criminal offender population (Williams, 1996). More importantly, it is considered essential that any rehabilitation programming being delivered to these types of offenders be theoretically sound, based on research, and provided in priority to those offenders who require them most (Gordon, Holden, & Leis, 1991). Nevertheless, a dilemma remains in terms of deter-mining what risk management model works best and for whom it may be most effective.

Safe reintegration

Of all the factors that influence public safety, criminal justice system service providers in collaboration with releasing authorities, can affect the safe release of offenders into the community. There is solid evidence supporting the premise that the gradual and structured release of offenders is the safest strategy for the protection of society against new offences by released offenders. For example, recidivism studies (Waller, 1974; Harman & Hann, 1986) have found that the percentage of safe returns to the community is higher for supervised offenders than those released with no supervision. Therefore, reintegration is seen as working to better prepare offenders for release and providing them with greater support once they are in the community. Reintegration efforts should yield dividends in terms of higher rates of safe return to the community and lower rates of criminal recidivism.

RISK MANAGEMENT

The public is very concerned with the manner in which criminal offenders are managed because those providing reintegration services are seen as being responsible for their safety. In keeping with this important task, Motiuk (1995, p. 24) notes:

Faced with the fact that most offenders eventually return to the community the best way to serve the public is to recognize the risk presented by an individual, and to then put to good use the tools, the training and our fundamental understanding of what it really means to manage offender risk. Effective risk management implies that decisions impacting on the organization are made using the best procedures available, are in keeping with the overall goals of the system.

For correctional service providers, the application of risk management principles to reducing the chance of criminal recidivism is all that is required to develop an effective risk management program (or to improve on an already existing one). These risk management principles include the assessment of risk; the sharing of information (communication); the monitoring of activities (evaluation); and if deemed appropriate, an intervention (incapacitation, programming). Public safety is improved whenever these risk management activities are integrated into every function and level of the organization providing care and control.

Many jurisdictions have been implementing new and improved offender risk assessment and management technology. The rest of the chapter addresses three important and related questions: “What are correctional outcomes?” and “How do we measure them?” Then, we ask ourselves a final question, “What else do we need to know?”

WHAT ARE CORRECTIONAL OUTCOMES?

In the criminological literature, there have been many attempts to show the relative efficacy of risk management procedures in meeting various correctional objectives. So far, attention has focused on both institutional adjustment and post-discharge/release outcome as the variables most relevant to criminal justice and mental health decision-making (Motiuk, 1991).

Most investigations exploring the issue of institutional adjustment have evaluated offenders in terms of disruptive or rule-breaking behaviour such as: riots, assaults, homicides, rule infractions, incident reports, misconducts, drug abuse, escapes, transfers, self-mutilations and suicides. Another large collection of investigations examining the topic of institutional adjustment has assessed offenders with respect to illness behaviour. For these studies, adjustment criteria have included illness complaints, sick call attendance, medical diagnosis, medication line attendance and hospitalizations.

Traditionally, studies addressing the topic of post-discharge/ release outcome have evaluated released offenders in terms of recidivism measures. The most significant of these measures have been arrest, reconviction, parole violation and return to prison. From the public's perspective, violent or sexual recidivism is an important problem to address because of its detrimental impact on victims. Moreover, it provides an indication of the effectiveness of correctional interventions (Lipton, Martinson, & Wilkes, 1975; Sechrest, White, & Brown, 1979).

HOW DO WE MEASURE THEM?

Resolving uncertainty about decisions, after all due consideration of relevant risk factors, is the cornerstone of any effective risk management program. In practice, the analysis of offender risk should serve to structure much of the decision-making with respect to custody/security designations, temporary/conditional release, supervision requirements and program placement. Therefore, it is not surprising to find ongoing attempts to design, develop and implement objective procedures for classifying offenders.

It is believed that comprehensive assessment at the intake/ admission stage is critical to the ability to gauge accurately risk during the later phases of the sentence, when decisions as to possible release are taken. At the same time, it is noteworthy that there are successful models of risk assessment for conditionally released offenders in the community. Such work can and has laid the foundation for developing assessment processes for all offenders at the front-end. The amalgamation of front-end and back-end processes into one integrated system requires the ability to conduct systematic and objective assessments upon intake/admission and to link up in meaningful ways (i.e., use the same language and cues) with community-based re-assessments.

Risk principle considerations address the assessment of risk, the prediction of recidivism, and the matching of levels of treatment service to the risk level of the offender (Andrews, Bonta, & Hoge, 1990). While there is considerable empirical evidence to support the “risk principle”, it cannot be made fully operational until a framework is put into place for establishing program priorities, implementing programs and allocating resources to best meet the needs of offenders.

An example: Sex offender treatment

The treatment of sex offenders is viewed as a therapeutic and structured intervention aimed at the reduction of risk to reoffend sexually (Motiuk, 1999). Although treatment may be more likely to reduce recidivism among higher risk sex offenders than their lower risk counterparts (Andrews & Bonta, 1994; Nicholaichuk, 1996), their higher risk level suggests that some of them will reoffend -- even after treatment. Unfortunately, the public is not likely to be very impressed with statistically significant treatment effects when some graduates from sex offender programs continue to reoffend (Gordon & Nicholaichuk, 1996). Nevertheless, some would argue that we still have a duty to reduce sex offences among higher risk sex offenders to prevent further victimization.

Reviewing the literature on the management and treatment of sex offenders is a formidable task. Especially, with the proliferation of published material devoted to the topic of sexual offending in recent years. However, one gets the impression of two groups headed in opposite directions -- on the one hand, policy makers with limited knowledge of psychology and risk prediction, and on the other hand, practitioners with limited understanding of crime and jurisprudence -- and arriving at very different conclusions from the same empirical evidence.

The sex offender treatment literature consists of a diverse collection of studies on exhibitionists, rapists, hebophiles, paedophiles, child molesters and incest offenders, sometimes subsumed under the general category of sex offender. The degree to which this heterogeneous group of sexual offence “subtypes” overlaps in treatment studies is difficult to determine. Where examinations of programs targeting discrete typologies or subtypes of sex offenders have produced some intriguing findings (Hagan, King, & Patros, 1994; Knight & Prentky, 1990; Lang, Pugh, & Langevin, 1988; Marshall & Barbaree, 1990), implications for the general sex offender population may be limited.

Sex offender programs have within them a diversity of treatment goals. These include: minimization and rationalization (Barbaree, 1991), attitudes and cognitive distortions (Murphy, 1990), social competence skills (Stermac & Quinsey, 1986), deviant arousal and fantasy (Laws & Marshall, 1990: Quinsey & Earls, 1990), anger management/impulse control (Prentky & Knight, 1986) and relapse prevention (Pithers, 1990). Unfortunately, some aggregations of treatment outcome position them all under the general heading of sex offender treatment.

Also important to consider is that sex offender treatment is conducted in different settings (residential, outpatient) with varying levels of intensity (duration, focus), employing different treatment techniques (cognitive-behavioural, pharmacological, psychotherapeutic) and modalities (individual, group). Consequently, any thorough review of the offender treatment literature would likely yield varying and inconsistent results (Lipsey, 1995).

Another reason for diverse findings in the literature is that many studies are characterized by the use (or selection) of heterogeneous offender samples, groups defined on the basis of rather loose criteria, and inappropriate control or comparison groups lacking fundamental matching procedures (Baxter, Motiuk, & Fortin, 1995). For example, in different treatment studies, participants have been identified as sex offenders on the basis of type of conviction, sexual preference, or measures of deviant sexual arousal. As well, comparing deniers or treatment dropouts to those who admit their sex offences or complete treatment, or incarcerated sex offenders to those on probation, may hold little potential for advancing our knowledge about treating sexual offenders. The absence of uniformity in operational definitions makes comparing across studies difficult, as it not at all clear that different treatment studies are examining the same or even a similar sex offender population.

For corrections, the random assignment of sex offenders to either “treated” or “non-treated” groups is especially problematic. While some sex offenders who are not motivated to receive treatment willingly do not participate, many service providers question the ethics of denying programming to sex offenders who wish to participate but do not for evaluative purposes (Marshall & Barbaree, 1990).

Examinations of sex offender treatment effectiveness examination should match treated offenders with untreated offenders on a set of relevant characteristics such as: being similarly situated, release date, age at release and sentence length (Motiuk, Smiley, & Blanchette, 1996). Ideally, the control or comparison group would also be matched with the treated group on risk factors such as: history of sexual offending and victim age/gender preferences. Such risk factors have been found to be significantly related to reoffending among sex offenders (Hanson & Bussière, 1996). This presents yet another methodological obstacle to over-come, as selection criteria for treatment (or exclusion) could have adverse impacts on the ability to conduct matching procedures.

Other methodological problems have arisen with using variable follow-up periods and different outcome measures. The exploration of sex offender recidivism, its correlates, and the impact of treatment in reducing the likelihood of its occurrence is a veritable challenge for any researcher. Explaining the causes and correlates of sexual reoffending is complicated by time at risk in the community (longer post-release periods necessarily results in greater higher recidivism rates), intensity of supervision, and a variety of moderating variables. Post-release outcome studies rarely concur on recidivism rates, partly due to the varying definitions of what constitutes “recidivism” (Freeman-Longo & Knopp, 1992). Treatment outcome measures include self-reports of new offences, charges, convictions, or returns to custody. On the other hand, more stringent definitions consider only new convictions for sex crimes as an outcome measure. Again, the absence of uniformity in measures makes comparing across studies difficult.

Interpretation of sex offender treatment studies is beset with a host of additional problems. Because of low base rates of sexual reoffending (Hanson & Bussière, 1996), sample sizes need to be exceedingly large (Marshall & Pithers, 1994). Moreover, reliance on officially recorded convictions may underestimate actual sexual recidivism rates. It is possible that a large amount of sexual offending remains undetected by these sources (Weinrott & Saylor, 1991). The problem is further compounded by sample attrition where individuals are removed from the treatment study or follow-up for a variety of reasons (Blanchette, 1996). Other methodological problems include detailing the therapeutic intervention under investigation, measurement of the service provider's adherence to the treatment protocol, and factoring in the delay between treatment completion and release.

These issues (heterogeneity of the sex offender population; differences in treatment goals, setting, intensity, technique and modality; selection of participants and non-participants; random assignment, matching, problems of definition; measurement of outcome) permeate the sex offender treatment literature to such an extent that synthesis of the major findings is often quite difficult.

While investigators may question the effectiveness of sex offender treatment to reduce sexual reoffending over extended time periods, the challenge is generally made on the basis that virtually all sex offender treatment outcome studies have methodological problems (Quinsey, Harris, Rice, & Lalumière, 1993). However, others have found that some treatments can be empirically demonstrated to be effective with sex offenders and are, in fact, successful in reducing sexual reoffending (Barbaree, Seto, & Maric, 1996; Marshall, 1996; Robinson, 1996).

In 1995, Hall conducted a meta-analysis of available sex offender treatment studies that showed a small, but robust, effect size for sex offender treatment. More specifically, Hall (1995) found that across studies, sexual recidivism for untreated sex offenders was 27%, compared with 19% for treated sex offenders. Therefore, on average, sex offender treatment tends to reduce sexual recidivism by approximately 30 percentage points. Similarly, others are reporting substantial reductions in sexual recidivism relative to comparison groups using a cognitive-behavioural approach of 24% (Gordon & Nicholaichuk, 1996).

In a multi-year multi-modal review of sex offender programs in Canadian federal corrections (Motiuk, 1998), a three-year follow-up of 210 treated sex offenders showed a 50% reduction in sexual recidivism (from 6% for the benchmark group to 3% in the program group). Similarly, Looman, Abracen, and Nicholaichuk (2000) explored long-term recidivism (average time at risk was 9.9 years) among treated and released sexual offenders from a Regional Treatment Centre and matched controls and found that the treated group had a reduction in sexual recidivism of 54%.

Albeit that attributing monetary value to human pain and suffering as well as life is controversial, a Criminology Research Council in Australia funded a recent study that investigated the economic costs and benefits of implementing prison-based sex offender treatment programs for male child sex offenders. Although cost-benefit analysis is always based on many assumptions, Donato and Shanahan (1999) estimate that, “if a 14 percentage point reduction in recidivism is achieved following an in-prison treatment program, this could result in an economic gain of up to $39,870 per prisoner, or $3.98 million for 100 treated prisoners”.

In a study of cognitive skills programming, Robinson (1996) reported a 58% reduction in the general recidivism of sex offenders who completed treatment while in prison. Although sex offenders appeared to achieve the greatest treatment gains relative to other offence groups (violent, drug, property) from cognitive skills training, about one-third had received sex offender treatment before participating in cognitive skills training. Here the question becomes one of whether the reduction in reoffending among sex offenders is attributable to sex offender treatment, cognitive skills training, or some combination thereof. Consequently, another methodological problem arises in the form of sequencing where sex offenders may have received more than one treatment before discharge. Future research into sex offender program effectiveness will undoubtedly have to address this issue.

Although there is some convergence among studies on the effectiveness of sex offender treatment in reducing sexual recidivism, treatment is not a unitary concept. Often, sex offenders are required to complete a variety of programs before being considered for release. Then, they may be required to participate in maintenance programs upon being cascaded in security level or placed in the community. As yet, the full impact or relative contribution of post-program efforts (i.e., relapse prevention) to reducing recidivism among sex offenders remains largely untested (Miner, Marques, Day, & Nelson, 1990).

The fact that sex offenders appear to be benefiting from treatment and that sex offenders are often required to complete pro-grams before discharge or release points to importance of continuing to offer specialized services to these individuals. It also emphasizes that research into sex offender program effectiveness must look deeper into the various components of a pro-gram before drawing hasty conclusions as to whether a particular treatment has had any impact. Nevertheless, a broader look at an overall system's impact on reducing repeat sexual offending can be expressed as follows.

A broader perspective ...

From the public's perspective, criminal recidivism is an important problem because it may provide an indication of the ineffectiveness of correctional interventions (such as probation, incarceration and treatment). A December 31, 2000, review of the Correctional Service of Canada Offender management system identified 3,428 sex offenders under federal jurisdiction, which accounts for about 16% of the total federal offender population (Motiuk & Vuong, 2001). This end-of-2000 review also determined that 66% of the sex offenders were incarcerated in federal institutions and 34% were being supervised on conditional release.

As noted earlier, an indices of repeat sex offending often used is the number of new charges recorded for released offenders during a particular time period. As noted in the Departmental Performance Report (Correctional Service of Canada, 2000), for all released federal offenders over a five-year period, from 1994-1995 to 1999-2000, the number of charges (not convictions) for sex assault decreased from 49 to 23, or by 47% (see Table 23.1). A note of caution is warranted here as “charges” may inflate the rate of recidivism as offences could be later cleared for various reasons.

TABLE 23.1 Charges for Sex Assault for All Released Federal Offenders (1994 to 2000)

.
 
1994-1995
1995-1996
1996-1997
1997-1998
1998-1999
1999- 2000
.
#
49
22
31
23
32
23
.
Source: Departmental Performance Report, Correctional Service Canada, 2000

Another indices of repeat sex offending often used is the number of crimes reported to police over a particular time period. As noted in the Uniform Crime Report, (Statistics Canada, 1999), for Canada, the actual number of sexual assaults (level 1, 2-with weapon, 3-aggravated) reported to police decreased from 31,706 to 23,872 or by 25% (see Table 23.2).

TABLE 23.2 Sex Assault Incidents Reported to Police in Canada (1994 to 1999)

.
 
1994
1995
1996
1997
1998
1999
.
Actual #
31,706
28,234
27,026
27,063
25,493
23,872
Rate/100,000
109
96
91
90
84
78
.
Adult males charged
10,434
9,062
8,498
7,847
7,887
7,361
.
Source: Uniform Crime Report, Statistics Canada, 1999

Also noteworthy, the rate of sexual assaults reported to police declined from 109 per 100,000 Canadian population in 1994 to 78 per 100,000 in 1999 and the number of adult males charged declined from 10,434 to 7,361 or by 30%. According to Correctional Service of Canada's Offender Management System and the 1999 Uniform Crime Report survey, there were 15 sex offences committed by federal offenders while on conditional release in 1999 and, therefore, as a group, accounted for 0.6 of every 1,000 sex offences reported in Canada that year. (Source: The Safe Return of Offenders to the Community, November 2000, Correctional Service of Canada).

However, an important question still remains. How many federal sex offenders who had completed specialized treatment commit new sex crimes while on conditional release or after their sentence is completed? To answer this question one can look to the literature. Between 1991 and 1994, an early study (Motiuk & Brown, 1993) found that among 570 released federal sex offenders (treated and untreated, with an average follow-up of 3.5 years), less than 1-in-10 (or 7%) were convicted for a new sexual offence. Similarly, Nicholaichuk (1996) reported a 6% sexual recidivism rate for a treatment group released from federal prisons.

Taken together, the average rate of sexual recidivism (6-7 % across federal studies) among released federal sex offenders is considerably lower relative to other published studies (average 13% -- see Hanson & Bussière, 1996). About half the rate found in other studies. In a more recent study (Motiuk, 1998), between 1994 and 1998, among treated and released federal sex offenders (with an average follow-up of 3.5 years), less than 1-in-33 were convicted for a new sex offence. The average rate of sexual recidivism (3.3%) among treated and released federal sex offenders is considerably lower relative to all released federal sex offenders. Again, we find about half the rate of all federal sex offenders.

WHAT ELSE DO WE NEED TO KNOW?

What constitutes recidivism? Is it is failure? ... returns to custody, ... technical violations, or ... new offences. How do we define and measure new offences? We seem to run into even more problems when we try to evaluate correctional success. Although “what is the recidivism rate” is a popular and valid question, it is really a difficult one to answer, and to emphasize any one answer can be misleading if we don't recognize its limitations.

A review of the available literature would reveal that the most common definition of recidivism is the percentage of released offenders returned to a correctional authority for a new offence during a particular period of study (Nouwens, Motiuk, & Boe, 1993).

Number of releases

Deciding on how to determine the number of released offenders raises several options and necessarily affects the denominator. For example, calculations may be used for the following:

  1. any release (under supervision and upon completion sentence),
  2. release under supervision (whether discretionary release -- parole or a presumptive release -- statutory release), and
  3. the aforementioned (flow) combined with those already under community supervision (stock) to complete the full picture of community supervision caseloads.

Naturally, the later is the basis on which most correctional systems would prefer to use as it reflects the full magnitude of case management effort required to reduce the likelihood of criminal recidivism.

Number of readmitted offenders

Deciding on how to determine the number of readmitted offenders also poses some choices and necessarily affects the numerator. For example, calculations may be used for the following:

  1. any return (under suspension, revocation, or new offence),
  2. return for technical violations of conditions, and
  3. return for a new offence (any, violent, or a specific offence like a sex offence).

Of course, the later is the basis on which most correctional systems prefer to measure recidivism as it reflects true relapse into crime.

Period of study

Deciding on how to determine the period of study has a few choices and necessarily affects both the numerator and denominator. For example, calculation may be used for the following:

  1. while under status (under sentence, post-sentence or both),
  2. a specified time frame (6 months, 1 year, 2 years, 3 years, 10 years, etc.), and
  3. a regular basis (annually).

Usually, the later is the basis on which most correctional systems prefer to measure recidivism as it reflects both recent and fiscally relevant accountability.

Even if we know the recidivism rate, we still can't be sure what it means and what accounts for it. We run into particular problems when we try to evaluate the success of correctional systems or pro-grams. Is the program successful if offenders who participated in it no longer commit offences related to the problem addressed by the program? For example, in 1998, we saw that the Correctional Service of Canada examined the post-release performance of over 1,000 treated sex offenders for an average of 3.5-year period and found that 17 cases recidivated sexually. Optimizing for length of follow-up, the recidivism rate rose to 3%, roughly half that of the entire sex offender when released. Overall, a good correctional result was obtained and whether it can be attributed in whole, or in part to prison-based sex offender treatment or in combination with effective community supervision practices may matter little in the long run. This result does, however, continue to pose methodological concerns for program evaluation purposes whenever recidivism rates are low.

SUMMARY

For evaluating correctional programs, reporting the change and reduction in recidivism for completers, participants and dropouts has become common practice. The change and reduction (reported as the difference in recidivism rate over the comparison group -- which raises the overall magnitude of the effect) in recidivism is measured relative to a matched comparison group, control group (sometimes program waiting list controls) or general base rate of recidivism for a similarly situated correctional population.

For evaluating correctional system(s) performance, reporting recidivism rates over time or comparing with other jurisdictions is also popular. A note of caution is warranted in conducting comparisons with others. Correctional systems exist in different countries with different social, political and judicial systems. Nevertheless, there continues to be sustained efforts to refine outcome definitions as well as optimism for new and better measures.


1 Correctional Service of Canada
The opinions expressed are those of the author and do not necessarily represent the official views of the Correctional Service of Canada or the Solicitor General Canada.

REFERENCES

Andrews, D. A. (1995). The psychology of criminal conduct and effective treatment. In J. McGuire (Ed.). What works? Reducing reoffending. (pp. 35-62). New York, NY: Wiley.

Andrews, D. A. (1996). Recidivism is predictable and can be influenced: An update. Forum on Corrections Research, 8(3), 42-44.

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

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

Barbaree, H. E. (1991). Denial and minimization among sex offenders: Assessment and treatment outcome. Forum on Corrections Research, 3(4) 30-33.

Barbaree, H. E., Seto, M. C., & Maric, A. (1996). Effective sex offender treatment: The Warkworth Sexual Behaviour Clinic. Forum on Corrections Research, 8(3), 13-15.

Baxter, D. J., Motiuk, L. L., & Fortin, S. (1995). Intelligence and personality in criminal offenders. In D. H. Saklofske & M. Zeidner ( Eds.), International Handbook of Personality and Intelligence, New York, NY: Plenum.

Blanchette, K. (1996). Sex offender assessment, treatment and recidivism: A literature review. Research Report R-48. Ottawa, ON: Correctional Service of Canada.

Bonta, J., & Motiuk, L. L. (1996). High-risk violent offenders in Canada. Research Report R-50. Ottawa, ON: Correctional Service Canada.

Donato, R., & Shanahan, M. (1999). The economic of implementing intensive in-prison sex offender treatment programs. Australian Institute of Criminology: Trends and Issues, 134.

Freeman-Longo, R. E., & Knopp, H. F. (1992). State-of-the-art sex offender treatment: Outcome and issues. Annals of Sex Research, 5(3), 141-160.

Gaes, G., Flanigan, T., Motiuk, L. L., & Stewart, L. (1999). Adult correctional treatment. In M. Tonry & J. Petersilia (Eds), Prisons. Chicago. IL: University of Chicago Press.

Gendreau, P., & Goggin, C. (1996). Principles of effective correctional programming. Forum on Corrections Research, 8(3), 38-41.

Gendreau, P., Goggin, C., & Cullen, F. (1999). The effects of prison sentences on recidivism. Report to the Department of the Solicitor General of Canada.

Gordon, A., Holden, R., & Leis, T. (1991). Managing and treating sex offenders: Matching risk and need in programming. Forum on Corrections Research, 3(1), 7-11.

Gordon, A., & Nicholaichuk, T. (1996). Applying the risk principle to sex offender treatment. Forum on Corrections Research, 8(2), 36-38.

Hagan, M. P., King, R. P., & Patros, R. L. (1994). The efficacy of a serious sex offenders treatment program for adolescent rapists. International journal of Offender Therapy and Comparative Criminology, 38(2), 141-150.

Hall, G. C. (1995). Sex offender recidivism revisited: A meta-analysis of recent treatment studies. Journal of Consulting and Clinical Psychology, 63(5), 802-809.

Hanson, R. K., & Bussière, M. T. (1996). Sex offender risk predictors: A summary of research results. Forum on Corrections Research, 8(2), 10-12.

Harman, W. G., & Hann, R. G. (1986). Predicting general release risk for penitentiary inmates. User Report: Ministry of the Solicitor General of Canada, No. 1986-32.

Knight, R. A., & Prentky, R. A. (1990). Classifying sex offenders: The development and corroboration of taxonomic models. In W. L. Marshall & H. E. Barbaree (Eds.), Handbook of Sexual Assault: Issues, Theories, and Treatment of the Offenders (pp. 23-52). New York, NY: Plenum.

Lang, R. A., Pugh, G. M., & Langevin, R. (1988). Treatment of incest and pedophilic offenders: A pilot study. Behavioral Sciences and the Law, 6, 239-255.

Laws, H. E., & Marshall, W. L. (1990). A conditioning theory of the etiology and maintenance of deviant sexual preference and behavior. In W. L. Marshall & H. E. Barbaree (Eds.), Handbook of sexual assault: Issues, theories, and treatment of the offenders (pp. 209-229). New York, NY: Plenum.

Lipsey, M. (1995). What do we learn from 400 research studies on the effectiveness of treatment with juvenile delinquents? In J. McGuire (Ed.). What works? Reducing reoffending. (pp. 35-62). New York, NY: Wiley.

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

Looman, J., Abracen, J., & Nicholaichuk, T. (2000). Recidivism among treated sexual offenders and matched controls. Journal of Interpersonal Violence, 15, 279-290.

Lösel, F. (1995). The efficacy of correctional treatment: A review and synthesis of meta-evaluations. In J. McGuire (Ed.). What works? Reducing reoffending. (pp. 79-114). New York, NY: Wiley.

Lösel, F. (1996). Effective correctional programming: What empirical research tells and what it doesn't. Forum on Corrections Research, 8(3), 33-37.

Marshall, W. L. (1996). Assessment, treatment, and theorizing about sex offenders: Developments during the past twenty years and future directions. Criminal Justice and Behavior, 23(1), 162-199.

Marshall, W. L., & Barbaree, H. E. (1990). Outcome of comprehensive cognitive-behavioral treatment programs. In W. L. Marshall, D. R. Laws, & H. E. Barbaree (Eds.), Handbook of sexual assault: Issues, theories, and treatment of the offender (pp. 363-385). New York, NY: Plenum Press.

Marsall, W. L., & Pithers, W. D. (1994). A reconsideration of treatment outcome with sex offenders. Criminal Justice and Behavior, 21, 10-27.

McGuire, J. (1995). What works? Reducing reoffending. New York, NY: Wiley.

Miner, M. H., Marques, J. K., Day, D. M., & Nelson, C. (1990). Impact of relapse prevention in treating sex offenders: Preliminary findings. Annals of Sex Research, 3(2), 165-185.

Motiuk, L. L. (1991). The antecedents and consequences of prison adjustment: A systematic assessment and reassessment approach. Doctoral dissertation. Ottawa, ON: Carleton University.

Motiuk, L. L. (1995). Refocusing the role of psychology in risk management: Assessment, communication, monitoring and intervention. In T. A. Leis, L. L. Motiuk, & J. R. Ogloff (Eds.), Forensic psychology: Policy and practice in corrections. Ottawa, ON: Correctional Service Canada.

Motiuk, L. L. (1996). Assessing sex offenders for responsivity. A published paper presented at the 15th Meeting of the Association for the Treatment of Sex Abusers, Chicago, Illinois.

Motiuk, L. L. (1997). Sex offender assessment: A risk-based case differentiation approach. A paper presented at the 16th Annual Research and Treatment Conference for the Association for the Treatment of Sex Abusers. Arlington, Virginia.

Motiuk, L. L. (1998). A multi-year multi-modal review of sex offender programs in federal corrections. A paper presented at the 17th Annual Research and Treatment Conference for the Association for the Treatment of Sex Abusers. Vancouver, British Columbia.

Motiuk, L. L. (1999). The treatment of sex offenders in penal establishments and release programs. A paper published in the proceedings of the 12th Conference of Directors of Prison Administration, Strasbourg, France.

Motiuk, L. L. (2001). Public administration and management of adult correctional services in Canada. In J. Winterdyk (Ed), Corrections in Canada: Social reactions to crime. Toronto, ON: Prentice-Hall.

Motiuk, L. L., Belcourt, R., & Bonta, J. (1995). Managing high-risk offenders: A post-detention follow-up. Research Report R-39. Ottawa, ON: Correctional Service Canada.

Motiuk, L. L., & Brown, S. L. (1993). Survival time until suspension for sex offenders on conditional release. Research Report R-31. Ottawa, ON: Correctional Service Canada.

Motiuk, L. L., Smiley, C., & Blanchette, K. (1996). Intensive programming for violent offenders: A comparative investigation. Forum on Corrections Research, 8(3), 6-9.

Motiuk, L. L., & Vuong, B. (2001). Homicide, robbery, sex and drug offenders in federal corrections: An end-of-200 review. Research Brief B-25. Ottawa, ON: Correctional service of Canada

Murphy, W. D. (1990). Assessment and modification of cognitive distortions in sex offenders. In W. L. Marshall, D. R. Laws, & H. E. Barbaree (Eds.), Handbook of sexual assault: Issues, theories, and treatment of the offender (pp. 331-342). New York, NY: Plenum Press.

Nicholaichuk, T. P. (1996). Sex offender treatment priority: An illustration of the risk/need principle. Forum on Corrections Research, 8(2), 30-32.

Nouwens, T., Motiuk, L., & Boe, R. (1993) So you want to know the redicivism rate. Forum on Corrections Research, 5(3), 22-26.

Nuffield, J. (1982). Parole decision-making in Canada: Research towards decision guidelines. Ottawa, ON: Solicitor General of Canada.

Otto, R. K. (1992). Measuring patient violence in dangerousness research. Law and Human Behavior, 17, 277-288.

Pithers, W. D. (1990). Relapse prevention for sexual aggressors: A method for maintaining therapeutic gain and enhancing external supervision. In W. L. Marshall, D. R. Laws, & H. E. Barbaree (Eds.), Handbook of sexual assault: Issues, theories, and treatment of the offender (pp. 343-361). New York, NY: Plenum Press.

Prentky, R. A., & Knight, R. (1986). Impulsivity in lifestyle and criminal behavior of sexual offenders. Criminal Justice and Behavior, 13, 141-164.

Quinsey, V. L., & Earls, C. M. (1990). The modification of sexual preferences. In W. L. Marshall, D. R. Laws, & H. E. Barbaree (Eds.), Handbook of sexual assault: Issues, theories, and treatment of the offender (pp. 279-295). New York, NY: Plenum Press.

Quinsey, V. L., Harris, G. T., Rice, M. E., & Lalumière, M. L. (1993). Assessing treatment efficacy in outcome studies of sex offenders. Journal of Interpersonal Violence, 8(4), 512-523.

Quinsey, V. L., Harris, G. T., Rice, M. E., & Cormier, C. A. (1998). Violent offenders: Appraising risk and managing risk. Baltimore, MD: United Book Press.

Robinson, D. (1996). Factors influencing the effectiveness of cognitive skills training. Forum on Corrections Research, 8(3), 6-9.

Ross, R. R., & Fabiano, E. (1985). Time to think: A cognitive model of delinquency prevention and offender rehabilitation. Johnson City, TN: Institute of Social Sciences and Arts.

Sechrest, L., White, S., & Brown, E. (1979). The rehabilitation of criminal offenders. Washington, DC: National Academy of Sciences.

Stermac, L. E., & Quinsey, V. L. (1986). Social competence among rapists. Behavioral Scientist, 8, 171-285.

Tarling, R. (1993). Analysing offending: Data, models and interpretations. London, UK: HMSO.

Waller, I. (1974). Men released from prison. Toronto, ON: Centre of Criminology, University of Toronto Press.

Weinrott, M. R., & Saylor, M. (1991). Self-report of crimes committed by sex offenders. Journal of Interpersonal Violence, 6(3), 286-300.

Williams, S. M. (1996). A National Strategy for Managing Sex Offenders. Forum on Corrections Research, 8(2), 33-35.

--------------------

Previous PageTop Of Page Table Of ContentsNext Page