FORUM on Corrections Research

Where Are We in Our Ability to Assess Risk?

In practice, the assessment of offender risk serves to structure many of the decisions we make regarding custody or security designations, temporary and conditional release, supervision requirements and program placement. The cornerstone of any effective risk management program is making decisions after all available information has been considered.

However, the capacity to conduct formalized risk assessments is directly related to the resources a correctional agency has at its disposal. It is not surprising, therefore, that objective assessment procedures for classifying criminal offenders have spread throughout North America.⁽¹⁾

The Past

Most assessment instruments being used today were originally crafted during the late 1970s and early 1980s. Some of these tools include the Level of Supervision Inventory (LSI);⁽²⁾ the Wisconsin Assessment of Client Risk Scale;⁽³⁾ the Psychopathy Checklist (PCL);⁽⁴⁾ the Salient Factor Score (SFS);⁽⁵⁾ the Minnesota Multiphasic Personality Inventory (MMPI)-based Typology;⁽⁶⁾ and the Statistical Information on Recidivism (SIR) Scale.⁽⁷⁾

Although all these instruments use objective scoring techniques and scientific approaches, their acceptance into everyday correctional practice has been met with disdain, uneven implementation and, often, abandonment altogether. It seems that as soon as the crafters of assessment tools depart, distance themselves or shift their research focus, the tools they developed run the risk of being placed on the shelf.

The Present

So where are we in our ability to assess risk? The short answer to that question is: better off than we were a decade ago. Some crafters revised their instruments to improve our ability to both understand and predict criminal behaviour.⁽⁸⁾

However, if one simply asks "By how much are we now able to reduce the uncertainty of correctional outcomes using these tools?", the reply is somewhat less encouraging. We can make better predictions using any one of these risk instruments, but the amount of variance left unexplained still outweighs that which can be explained.

While this may be cause enough for disillusionment, it suggests that the next generation of risk assessors in corrections will have to view assessment as an integrated process incorporating a variety of methodologies. So, if one asks instead, "What did we learn from using these tools?", then maybe we can reserve some optimism for improved risk assessment in the near future.

Faced with the correctional challenges of the 1990s, we will need new assessment techniques (i.e., multi-method and multipredictor assessment) and the use of systematic reassessment⁽⁹⁾ to improve the way we manage risk. This situation is particularly true for the Correctional Service of Canada, which has recently undertaken the ambitious Correctional Strategy Initiative.⁽¹⁰⁾ This initiative puts into place a framework for establishing program priorities, implementing programs and allocating resources to meet the needs of offenders. Insofar as we have also placed a renewed emphasis on the safe reintegration of offenders, we have recognized the need for a comprehensive and integrated process to assess offenders upon admission to federal custody. A newly devised intake assessment process, designed and developed by the Correctional Service of Canada, exemplifies this new direction.

Before elaborating on the new model, it is important to reflect on the organization's recent breakthrough in implementing systematic offender risk/needs assessment in the community. This achievement has been instrumental in providing the conceptual foundation and impetus for an offender risk/needs assessment process upon entering federal custody. The strategy for risk management will be to conduct assessments upon admission to prison and link them up (i.e., use the same language and cues) with reassessments done during conditional release.

The Community Risk/Needs Management Scale

Research has found that criminal history factors are strongly related to outcome on conditional release; that a consistent relationship exists between the type and number of needs offenders present and the likelihood of their reoffending;⁽¹¹⁾ and, most important, that the combined assessment of both risk and needs levels significantly improves our ability to predict who is likely to reoffend and who is not.⁽¹²⁾

In October 1988, as part of the field testing of new standards for conditional release supervision,⁽¹³⁾ case management staff of the Correctional Service of Canada used a systematic approach to assess the needs of offenders, the risk of reoffending and any other factors that might affect successful reintegration into the community. A Community Risk/Needs Management Scale was designed, developed and implemented. It involves using case-specific information on criminal history and a critical set of case needs dimensions to classify federal offenders on conditional release⁽¹⁴⁾.

Criminal History Risk Assessment

To assess the risk of reoffending systematically and consistently, case managers use the SIR Scale, which has been officially adopted by the National Parole Board as a release-risk scoring system. The SIR Scale involves an extensive review of an individual's official criminal record. In addition, case managers use two other sources of criminal history information so the level of risk according to criminal history can be determined in an objective, reliable and accurate way. Case managers also use the National Parole Board's overall assessment of risk (i.e., low versus not low) and their own judgment of criminal history risk based on a thorough review of an offender's criminal record.

Case Needs Assessment

The needs areas selected for this component of the Community Risk/Needs Management Scale are typical of those included in most other needs assessment scales used in various jurisdictions.⁽¹⁵⁾ A total of 12 areas are covered: academic and vocational skills, employment pattern, financial management, marital and family relationships, companions and significant others, living arrangements, behavioural and emotional stability, alcohol usage, drug usage, mental ability, health and attitude. Although each area is rated according to specified guidelines, an overall rating of needs is given simply by compiling case manager judgments into one of three needs levels: low, medium or high.

The appropriate frequency of contact for parole supervision is determined by linking the two types of assessments - criminal history risk and case needs - in a matrix format (e.g., high risk/high needs). To ensure that the scale would meet the community supervision needs of certain special categories of offenders (i.e., sexual and mentally disordered), two additional special needs categories were included. A special needs category of "other" is reserved for those who do not meet the aforementioned criteria but are viewed by case managers as meriting a higher rating.

The field test of the Community Risk/Needs Management Scale found that, by simply combining case manager assessments of criminal history risk with global ratings of case needs, up to 47.5% of offenders who had been assessed as being in the high-risk/high-needs group were suspended within six months of their initial assessment. On the other hand, substantially fewer of those assessed as low-risk/low-needs offenders (5.1%) were suspended while on conditional release. Of particular interest, this low-risk/low-needs group was the largest category among the identified risk/needs-level groupings, representing 35% of the total sample of assessed cases.⁽¹⁶⁾ Therefore, reducing the frequency of supervision for these lower-risk cases had important implications for the reallocation and refocusing of community resources.⁽¹⁷⁾

Presently, the Community Risk/ Needs Management Scale is systematically administered and readministered to federal offenders under community supervision by case managers across Canada. We have developed a computerized means to monitor offender risk/needs levels by using the Offender Population Profile System (OPPS).18 Through OPPS, the overall risk/needs levels gathered since the implementation of the Community Risk/Needs Management Scale are being stored and can be retrieved to provide monthly snapshots.

Table 1 shows a national overview of the risk/needs levels of the conditional release population over the last two years. Have there been any significant changes in the profile of the conditional release population? The table illustrates a significant trend in the steady decline of the proportion of cases assessed as being low risk/low needs (31.6% to 27%) as opposed to the steady increase in the proportion of cases assessed as being high risk/high needs (25.3% to 36.1%).

---

Natioanl Overview of the Conditional Release Population: Percentage Distribution of Risk/Needs Levels
Risk/Needs Levels	Period
	December 1990 (7,023)	June 1991 (7,800)	December 1991 (8,189)	June 1992 (8,453)	December 1992 (8,666)
Low-Low	31.6 %	30.3 %	28.2 %	28.6 %	27.0 %
Low-Medium	26.0 %	24.9 %	24.9 %	26.0 %	24.9 %
Low-High	2.4 %	2.4 %	2.8 %	2.5 %	3.0 %
High-Low	3.0 %	2.7 %	2.2 %	1.6 %	1.5 %
High-Medium	11.8 %	9.5 %	6.9 %	7.8 %	7.5 %
High-High	26.3 %	30.3 %	33.2 %	35.4 %	36.1 %
Unassessed	1,739	1,298	942	664	752

Does this mean that the public is at greater risk? The definitive answer to this question remains unclear, as the changes may represent the drift in risk assessment over time. That is, case managers may be overestimating the level of risk. This may be somewhat akin to the practice of overclassification found in many institutional populations.⁽¹⁹⁾ What this information does tell us is how field staff have been responding to the conditional release population over time. For certain, if our frequency of contact guidelines are being strictly adhered to, then a substantially larger proportion of offenders on conditional release are being supervised much more closely now than ever before.

As expected, being able to produce an offender risk/needs profile of our entire conditional release population has been extremely useful for raising awareness about community supervision, providing basic statistics on risk/needs levels and estimating resource implications. This has moved the organization closer toward an effective risk management program. Now the question becomes: "Can we improve on this work?"

Offender Intake (Front-End) Assessment Process

In August 1991, under the auspices of the Correctional Strategy Initiative,⁽²⁰⁾ it was decided that an offender's needs should be the basis for programming, and that service delivery should focus primarily on successful reintegration into the community. However, the tool that was being used to conduct assessments upon admission, the Force-field Analysis of Needs, was inadequate for profiling offender needs. Consequently, a national working group drafted a new scheme to improve assessment of criminal risk and identify offender needs at the time of admission.

The development of the new instrumentation purposefully followed, and expanded on, existing assessment tools, namely the Case Management Assessment Interview, the Force-field Analysis of Needs and the Community Risk/Needs Management Scale. The aim was to capitalize on existing information-gathering practices, retain essential outputs (e.g., offender groupings) and build on case management training to date.

Case Needs Identification and Analysis

The new Case Needs Identification and Analysis protocol collapsed the 12 need areas of the Community Risk/Needs Management Scale into 7 need dimensions or target domains. These include: employment, marital and family, associates and social interaction, substance abuse, community functioning, personal and emotional orientation, and attitude. Following the initial version of the Case Needs Identification and Analysis protocol, the working group crafted the other assessment domains (see Table 2) for the Offender Intake (also known as Front-End) Assessment Process.

---

Offender Intake Assessment Domains

1) postsentence community assessment

2) initial assessment (mental health, security, suicide risk, etc.)

3) criminal risk assessment criminal risk assessment offences severity record sex offender history checklist detension criteria & SIR Scale

4) case needs identification and analysis employment marital & family associates & social interaction substance abuse community functioning personal & emotional orientation attitude

5) psychological assessment

6) supplementary assessment(s)

Rating guidelines are provided for each of the seven need dimensions. An overall needs rating is the compilation of professional judgments derived from the results of an initial assessment (medical, mental health, suicide risk) and the observations or impressions (i.e., degree or severity of need) on each of the seven need areas.

The Offender Intake Assessment Process⁽²¹⁾ represents the latest generation of risk assessment technology. It integrates information gathered from a variety of sources using many techniques. While the mechanics of the whole intake assessment process are beyond the scope of this paper, a closer look at the improvements we are making in criminal risk assessment may tell more about our ability to assess risk.

Criminal Risk Assessment

Upon admission to federal corrections institutions, every federal offender is rated for criminal risk based on the following: the criminal history record, the offence severity record, the sex offence history checklist, whether detention criteria are met, the result of the SIR Scale and any other risk factors described in a criminal profile report which provides details of the crime for which the offender is currently sentenced.

The Criminal History Record

By systematically reviewing the offender's file, which includes police reports, court transcripts and criminal records, a criminal history record is completed. Information is gathered on previous offences, current offences, the number and types of convictions, youth court dispositions, adult court sanctions and crime-free periods. This information is compiled into three separate indices of the criminal history record: previous youth court involvement, previous adult court involvement and current offence. Together, these yield a total score that reflects the nature and extent of an offender's involvement with the criminal justice system.

The Offence Severity Record

Similarly, a systematic review of the offender's file is used to complete an offence severity record. Information is gathered on current offences, previous offences, types of convictions, sentence length, the number and types of victims, the degree of force used on victims and the degree of physical and psychological harm to victims. This yields a total score that reflects the nature and degree to which an offender has inflicted harm on society in general, and on victims in particular.

The Sex Offender History Checklist

The offender's file is reviewed thoroughly to complete a sex offence history checklist. This checklist consists of the following: sex offender status, types of previous and current sex offences, victims, determination of serious harm to victims, assessment and treatment history, and a summary. Offenders are identified as sex offenders if they are serving a sentence for a sex offence, have been convicted in the past for one or more sex offences, are serving a sentence for a sex-related offence or have previously been convicted of an offence that is sex-related.

The types of current and past sex offence are identified as one or more of the following: incest, pedophilia, sexual assault and other sex offences (e.g., voyeurism, exhibitionism, fetishism, bestiality). With respect to victims, information on their number, gender and age is recorded. The determination of serious harm is based on whether the current offence resulted in death or serious harm. Information is also gathered on previous psychological or psychiatric assessments, earlier treatment or intervention and current treatment or intervention for sex offending.

All this information is compiled to yield a total score that reflects the nature and extent of sex offending, the amount of harm inflicted on victims and the involvement in assessment, treatment or intervention in relation to sex offending.

Detention Criteria and SIR Scale

A review of detention criteria for current offences reflects the nature of the offences and the degree of harm to victims. Then, the SIR Scale, a statistically derived tool for predicting recidivism, is completed. The scale combines measures of demographic characteristics and criminal history in a scoring system that estimates the chance of recidivism for different groups of offenders.

Rating Criminal Risk
An overall criminal risk rating of "low" is given when the criminal history record score reflects little or no involvement with the criminal justice system; when the offence severity record score reflects little or no harm to society in general, and to victims in particular; when a review of the sex offence history reflects little or no sexual offending; and when a review of the detention criteria and the SIR Scale support all of the aforementioned indices. This approach uses objective tools such as the SIR Scale in the risk assessment process to help make decisions on low risk.

Overall ratings for medium or high criminal risk are determined from a systematic review of professional judgments derived from the criminal history record, the offence severity record and the sex offence history components of the criminal risk assessment protocol. Offenders who are currently serving sentences for offences that caused death or serious harm must be given a high-risk rating. As well, an overall rating of high criminal risk is given to offenders assessed as having any of the following: a criminal history record score that reflects extensive involvement with the criminal justice system; an offence severity record score that reflects considerable harm to society in general, and to victims in particular; or a sex offence history that reflects considerable sex offending.

When the review of professional judgments concludes that the offender is clearly not a low criminal risk and there is sufficient reason not to rate the offender as a high risk, then an overall rating of medium criminal risk is deemed appropriate. Establishing criminal risk levels also incorporates a great deal of other assessment information. For example, additional information might be obtained from specialized assessments (e.g., of sex offenders) and case conferences.

The Link Between Assessment in Institutions and in the Community

The Offender Intake Assessment Process is being tested at six penitentiaries across Canada. while the case needs identification and analysis portion was developed principally for assessing offender needs upon admission to federal custody, an Ontario region correctional strategy group has streamlined the scope of needs assessment for community-based assessment. Because the group kept the individual ratings for the levels of both criminal risk and case needs as well as for each need area, it will be able to align its community version of risk/needs assessment with the intake version. Presently, testing of this community risk/needs assessment process is under way in all parole offices and private sector agencies in the Ontario region.

Offender Intake Assessment System

Of particular concern in the offender assessment system is a requirement to automate fully the Offender Intake Assessment Process. One of the modules planned for our new nation-wide computer-based Offender Management System (OMS) is an integrated Offender Intake Assessment System (OIAS). The OIAS provides a computer-based structure for systematically collecting and integrating the criminal risk and case needs rating.

The OIAS pilot system is being field-tested and developed as a stand-alone microcomputer application. Once development and testing are completed, this system will be put on the OMS national network.

The Future

It appears that the day is quickly approaching when we will be able to assess offenders upon admission to correctional facilities in a comprehensive, integrated and systematic fashion and routinely reassess them thereafter. OIAS technology can now be seen on the horizon. In addition, we are getting ready to use artificial intelligence-based technology to deliver an offender assessment tutoring system to assist staff.

Where are we in our ability to assess risk? We are about to make some important breakthroughs.

---

⁽¹⁾D.A. Andrews, J. Bonta and R. Hoge, "Classification for Effective Rehabilitation: Rediscovering Psychology," Criminal Justice and Behavior, 17(1990): 19-52.
⁽²⁾D.A. Andrews, The Level of Supervision Inventory (LSI): The First Follow-Up (Toronto: Ontario Ministry of Correctional Services, 1982).
⁽³⁾S.C. Baird, "Probation and Parole Classification: The Wisconsin Model," Corrections Today, 43 (1981): 36-41.
⁽⁴⁾R.D. Hare, "A Research Scale for the Assessment of Psychopathy in Criminal Populations," Personality and Individual Differences, 1 (1980): 111-117.
⁽⁵⁾P.B. Hoffman, "Screening for Risk: A Revised Salient Factor Score," Journal of Criminal Justice, 11 (1983): 539-547.
⁽⁶⁾E.L Megargee and M.J. Bohn, Classifying Criminal Offenders: A New System Based on the MMPI (Beverly Hills, Calif: Sage, 1979).
⁽⁷⁾J. Nuffield, Parole Decision-Making in Canada: Research Towards Decision Guidelines (Ottawa: Solicitor General of Canada, 1982).
⁽⁸⁾L.L. Motiuk, "Using the LSI and Other Classification Systems to Better Predict Halfway House Outcome," IARCA Journal on Community Corrections, 5 (1993): 12-13.
⁽⁹⁾L.L. Motiuk, Antecedents and Consequences of Prison Adjustment: A Systematic Assessment and Reassessment Approach. Ph.D. dissertation, Carleton University, 1991.
⁽¹⁰⁾Correctional Service Canada, "Proceedings of the Correctional Strategy Conference," (Ottawa, 1991).
⁽¹¹⁾J. Bonta and L.L. Motiuk, "Utilization of an Interview-Based Classification Instrument: A Study of Correctional Halfway Houses," Criminal Justice and Behavior, 12 (1985): 333-352.
⁽¹²⁾J. Bonta and L.L. Motiuk, "The Diversion of Incarcerated Offenders to Correctional Halfway Houses," Journal of Research in Crime and Delinquency, 24 (1987): 302-323.
⁽¹³⁾Correctional Service of Canada and National Parole Board, Standards for Conditional Release Supervision: Draft for Consultation (Ottawa: Correctional Service of Canada and National Parole Board, 1988).
⁽¹⁴⁾L.L. Motiuk, "Identifying and Assessing Needs of Offenders Under Community Supervision: The Conditional Release Supervision Standards Project" Paper presented at the First Annual Forum on Corrections Research, Ottawa, 1989.
⁽¹⁵⁾L.L. Motiuk and F.J. Porporino, "Offender Risk/Needs Assessment A Study of Conditional Releases," Report R-01 (Ottawa: Correctional Service of Canada, 1989).
⁽¹⁶⁾L.L. Motiuk and F.J. Porporino, "Field Test of the Community Risk/Needs Management Scale: A Study of Offenders on Caseload," Report R-06 (Ottawa: Correctional Service of Canada, 1989).
⁽¹⁷⁾L.L. Motiuk and J. Bonta, "Prediction and Matching in Corrections: An Examination of the Risk Principle in Case Classification." Paper presented at the Canadian Psychological Association Annual Convention, Calgary, 1991.
⁽¹⁸⁾L.L. Motiuk and R. Boe, "Using SAS Software to Deliver an Offender Population Profile System (OPPS)," Proceedings of the Seventeenth Annual SAS Users Group International Conference, Hawaii, 1992.
⁽¹⁹⁾J. Bonta and L.L. Motiuk, "Inmate Classification," Journal of Criminal Justice, 20(1992): 343-353.
⁽²⁰⁾Correctional Service Canada, "Proceedings of the Correctional Strategy Conference."
⁽²¹⁾L.L. Motiuk and D. Pisapio, "Correctional Strategy: Front-end Assessment." Paper presented at the Cognitive Living Skills Coaches Convention, Montreal, Quebec, 1992.