James Byrne, Guest Editor
Department of Criminal Justice and Criminology
University of Massachusetts, Lowell
Every day, federal, state, and local community corrections officers are required to make decisions about the “risk level” posed by individuals placed under community supervision. Based on this risk assessment, offenders are assigned to a specific supervision level, where a variety of “tools and techniques” will be applied in an attempt to “manage” the risk posed by these offenders, at least for their time under community supervision. With roughly 100,000 probation and parole officers nationwide supervising an offender population of close to 5 million, it can certainly be argued that risk assessment is the single most important decision made by probation and parole officers today. With large caseloads and shrinking budgets, “triage” is the “name of the game” in the community corrections field; and assessment is a critical first step in any triage process.
In this special issue of Federal Probation, a number of nationally known experts in the field of community corrections have been asked to provide their own assessments of this triage process, focusing specifically on the use of both clinical and actuarial risk assessments by community corrections officers. A number of questions about risk assessment are raised and answered in this issue, including:
- What do we actually mean by risk?
- How reliable and accurate are actuarial and clinical assessments?
- What types of assessment instruments are available for adult offenders, juvenile offenders, and specific offender groups, such as sex offenders, drug offenders, and/or mentally ill offenders?, and perhaps most importantly,
- What evidence exists that links initial (and ongoing) risk assessment (using actuarial and clinical assessment tools) with subsequent offender outcomes (i.e., the link between assessment and risk reduction)?
As the authors of the articles in this special issue demonstrate, we currently know more about how to classify offenders into several categories of “risk” of re-offending than we know about how to reduce their “risk” of re-offending while under community supervision. When viewed in this context, debates over the use of actuarial vs. clinical assessment tools tend to obscure a larger issue: Are we interested in short-term offender control or long-term offender change? I would argue that in an evidence-based community corrections system, a clear link needs to be established, beginning with 1) valid and reliable initial (and ongoing) assessment, continuing to 2) development of risk-specific community corrections interventions, and concluding with 3) identification of subsequent offender outcomes in communities with different community risk levels.
Because risk assessment is the “lynchpin” in this process, it is critical to the success of community corrections, both in terms of short-term offender control and long-term offender change. While much time and effort has focused on how to assess the risk level of individual offenders, far less research has been focused on the assessment of community risk level (see, e.g. Pattavina, et al., 2006). As we conduct further research on how different individuals (such as high- vs. low-risk offenders) respond to community supervision strategies in different communities (such as high- vs. low-risk neighborhoods), we will take the logical next step in the development of an evidence-based community corrections system.
In the first article of this special issue, “Assessment with a Flair: Offender Accountability in Supervision Plans,” Faye Taxman—one of this country’s leading experts on case planning in community corrections—argues that the first rule in evidence-based community corrections practice is that services (both treatment and control-based) to offenders should vary in intensity based on the risk level of offenders. She describes how the risk level of the offender should be considered: first in terms of the intensity and appropriateness of services, and next in terms of the offender’s individual case plan, which should focus on the clinically-based assessment of criminogenic needs (substance abuse, family dysfunction, peer associates, criminal personality, antisocial cognitions, low self control, and other factors, including mental health).
Patricia Harris expands on the notion that both actuarial and clinical assessments are essential components of an evidence-based community corrections system in her excellent review of both types of risk assessment in community corrections. Her article, “What Community Supervision Officers Need to Know About Actuarial Risk Assessment and Clinical Judgment,” reviews the empirical research on the adequacy of both clinical and actuarially-based risk assessment instruments, and then identifies three impediments to the full implementation of actuarial risk assessment in community corrections:
- unclear/cursory presentations on the purpose of actuarial risk assessments to line staff;
- poor communication of offender risk assessment results; and
- failure to recognize the importance of clinical judgment (and skill) in the (actuarial) risk assessment process.
The third article that focuses on the use of actuarial and clinical assessment tools—“Clinical versus Actuarial Judgments in Criminal Justice Decisions: Should One Replace the Other?”—is authored by two of the country’s foremost experts on risk assessment, Stephen Gottfredson and Laura Moriarty. They begin their review by offering the following unequivocal assessment of the two competing approaches to risk assessment: “In virtually all decision-making situations that have been studied, actuarially developed devices outperform human judgments.” While they argue that “there is a place for human judgment and experience in the decision-making process,” they offer the following caveat: over-reliance on human judgment may undermine the accuracy of the risk assessment, because probation and parole officers may “concentrate on information that is demonstrably not predictive of offender behavioral outcomes.”
While the three articles that have just been highlighted present an overview of existing research that has been conducted on both actuarial and clinical risk assessment instruments, the next three articles highlight the use of different types of assessment strategies for different types of offenders. First, Albert Roberts and Kimberly Bender examine the most commonly used risk and mental health needs assessment tools in juvenile corrections in their uniquely titled article, “Overcoming Sisyphus: Effective Prediction of Mental Health Disorders and Recidivism among Delinquents.” The authors discuss the implications of their review for the two central goals of youth assessment in juvenile justice settings: 1) the safety of the community, and 2) the rehabilitation (and clinical treatment) of individual juvenile offenders. [ Note: According to the “Myth of Sisyphus” webpage, “The gods had condemned Sisyphus to ceaselessly rolling a rock to the top of a mountain, whence the stone would fall back of its own weight. They had thought with some reason that there is no more dreadful punishment than futile and hopeless labor.” ]
Next, Lurigio and Swartz identify and critically review the use of actuarial and clinical risk assessment tools for the large number of adult offenders in our community corrections system with serious mental illness (SMI). Focusing specifically on the instruments used to identify offenders with serious mental illness, Lurigio and Swartz describe the results of their recent research testing the accuracy of two “new generation” mental health screening tools in community corrections (the K-6 screening tool, and the Brief Jail Mental Health Screen).
Moving from mentally ill offender assessment devices to sex offender risk assessment devices, Andrew Harris provides a comprehensive review of the research on the use of actuarially-based and clinically-based assessments of sex offender risk. Harris’s article underscores one limitation—at least to some—of a risk-driven community corrections system: If we made decisions about sex offenders based solely on actuarial risk assessment, then very few sex offenders would ever be placed under close supervision, because as a group, sex offenders recidivate at remarkably low rates. The Harris article raises important questions for community corrections managers to consider, not only about the accuracy of sex offender-specific assessment devices using either actuarial or clinical assessment instruments, but also about the interplay between the actual risk posed by various types of sex offenders and the stakes (for offenders, victims, probation/parole officers, and the community) associated with decisions made on the appropriate level of control needed for this group of low-risk but high-stakes offenders. For sex offenders, over-classification appears to be an inevitable consequence of the “risk” assessment process, not because these offenders pose a high risk to recidivate, but rather because for this group of offenders, any risk—one in 10, one in 20 or even 1 in 100—is unacceptable due to the high stakes involved.
One of the challenges facing community corrections managers at the federal, state, and local level is the development of a defensible risk assessment system. By defensible, I am referring to a classification system that is externally reviewed and objectively validated. Two separate articles address this issue directly. First, Anthony Flores, Christopher Lowenkamp, Paula Smith, and Edward Latessa examine the predictive accuracy of the Level of Service Inventory-Revised (LSI-R) for a sample of 2,107 adult federal probationers, using subsequent incarceration (rather than the traditional re-arrest) as their outcome measure. Next, Susan Turner and Terry Fain present the results of their validation of the “Risk and Resiliency Checklist,” first developed in San Diego and currently being used in the Los Angeles Probation Department as the centerpiece of the department’s new automated case planning system. The authors found that a youth’s resiliency score—the net sum of risk factors (which have negative values) and protective factors (which have positive values)—is significantly related to subsequent recidivism, using re-arrest during a 12-month follow-up period (from initial assessment) as the outcome measure.
The final two articles in this special issue challenge much of the current thinking in the area of risk assessment and offer a different set of policy options for readers to consider. James Austin is one of this country’s foremost authorities on institutional and community-based classification systems. His article, “How Much Risk Can We Take? The Mis-use of Risk Assessment in Corrections” describes the six basic steps to follow in developing a risk assessment instrument:
- Risk assessment instruments must be tested on your (own) correctional population and separately normed for males and females,
- An inter-rater reliability test must be conducted,
- A validity test must be conducted,
- The instrument must allow for dynamic and static factors,
- The instrument must be compatible with the skill level of your staff, and
- The risk instrument must have face validity and transparency with staff, probationers, parolees, and policy makers.
Utilizing the concepts presented in his six-step risk assessment development model, Austin provides a critique of the best-known and most commonly used risk assessment instrument, the LSI-R, using illustrative examples from recent studies on the reliability and validity of the LSI-R that he conducted in Pennsylvania (parole board), Washington, and Vermont (community corrections). Austin concludes his review by identifying the “disconnect” that currently exists at the federal, state, and local level between an offender’s risk level and the availability of services: high-risk offenders are under-serviced, while low-risk offenders are over-serviced. As Austin suggests, “It would be helpful for those in the risk assessment business to start advocating a more reasonable level of intervention that matches the risk they have so carefully calibrated” (this volume).
Byrne and Pattavina’s article, “Clinical and Actuarial Risk Assessment in an Evidence-based Corrections System: Issues to Consider,” concludes the special issue by identifying three topics central to the current debate over the use of actuarial and clinical risk assessment in community corrections: 1) the need to distinguish between risk assessment and risk reduction, 2) the dummying down of community corrections associated with the development of actuarial risk assessment instruments, and 3) the need to combine individual risk assessment and community risk assessment in the next generation of risk-driven supervision strategies.
Taken together, the ten articles included in this special issue of Federal Probation offer readers an opportunity to examine the empirical evidence on the application of actuarial and clinical assessment instruments in community corrections systems, and to consider the new wave of assessment instruments being developed for adult and juvenile offenders and for specific subgroups of offenders (mentally ill offenders, substance abusers, sex offenders). Each author raises challenging issues for policy makers, practitioners, and the public to consider regarding the proper role of assessment in an evidence-based community corrections system.
What “predictions” can be offered about the direction of the field? In the very near future, I suspect that the current focus on the reliability and validity of risk classification systems will be supplanted by discussions of how to improve treatment classification systems in both institutional and community corrections. When this occurs, it will likely be the result of an emphasis on offender change—rather than short-term offender control—as the primary purpose of our community corrections system.