The development and proliferation of offender risk assessment tools have allowed corrections professionals to assess an offender’s risk to reoffend and determine the needs underlying his or her criminality. Not only are risk tools the engine that drive effective interventions with offenders, but they help agencies be more efficient by helping them identify those offenders that need the most intensive intervention and that pose the most risk to reoffend.
Current risk assessment instruments help agencies identify who should receive the most supervision or intervention to reduce their risk of recidivism (i.e., higher risk offenders) and determine what specific factors should be targeted to change criminal behavior. These factors are called criminogenic needs (i.e., factors that are strongly correlated with criminal behavior and are dynamic in nature). Through years of study, researchers have identified the major criminogenic needs of offenders: (a) antisocial attitudes, values, and beliefs; (b) antisocial peers; (c) antisocial personality characteristics; (d) family; (e) education/ employment; (f) leisure/recreational activities; and (g) substance abuse. This entry provides a general overview of offender risk assessment practices and processes, reviews the history and current state-ofthe-art of risk assessment instruments, outlines some common practical and methodological issues related to their use and implementation, and examines related risk level classification issues.
What Are Offender Risk Assessment Tools?
The process of identifying an offender’s risk level and corresponding criminogenic needs is known as offender risk assessment or offender classification. Put simply, risk refers to an offender’s likelihood to recidivate or to get into trouble again. Thus, offenders who are at high risk are more likely to reoffend, compared to offenders who are at low risk. In a similar vein, research indicates that a reduction in an offender’s risk level is typically associated with a reduction in his or her likelihood of recidivism, while an increase in an offender’s risk level is typically associated with an increase in his or her likelihood of recidivism.
Commonly used instruments consist of actuarial risk assessments that are similar to what insurance companies use to calculate insurance rates. Car insurance rates, for example, are based on the statistical analysis of a person’s driving records and other supplemental information (e.g., age, where the driver lives, type of car he or she drives), which helps insurance companies develop probability tables based on group behavior. A probability table, for instance, may indicate that someone is a high-risk driver due to his or her history of car accidents and speeding tickets, where he or she currently resides, and the type of car he or she drives. Similarly, in the field of corrections, a risk assessment tool may indicate that someone is a high-risk offender due to his or her criminal history, use of illegal substances, and unemployment status. Ultimately, insurance companies and criminal justice professionals alike can use the results of actuarial risk assessments to determine who is riskiest to commit a new driving offense or a new criminal offense, respectively.
Actuarial-based instruments used in the field of corrections today typically consist of a series of items that focus on those factors that have been found to be related to criminal behavior (e.g., antisocial peers, substance abuse). Items on the tool are then statistically weighted based on how strongly they correlate with recidivism, and an overall score is calculated that identifies the individual’s risk to reoffend. The corresponding risk level and information related to the offender’s unique areas of need can then be used to help guide decisions about whether to sentence the offender to prison or probation, his or her conditions of supervision, and the types of resources that will be used, all while enhancing public safety and security.
History of Offender Risk Assessment
The first generation of risk assessments relied on corrections professionals’ judgment and experience (i.e., their gut feelings). While information was collected through a face-to-face interview and a review of the offender’s file, prediction about their likelihood to reoffend was made based on the assessor’s professional opinion. In this way, there was no structured or standardized set of questions to be asked, so the assessor could ask the offender as many or as few questions as the assessor deemed appropriate. Ultimately, decisions based on this approach were idiosyncratic, subjective, and biased. Furthermore, this approach did not yield information that resulted in a distinction between risk levels.
The second generation of formal classification instruments was first developed by Andrew Alexander Bruce and his colleagues in 1928 at the request of the Illinois Parole Board. The Illinois Parole Board, more specifically, wanted to make more informed decisions about whom to release and when to release individuals back into the community. As such, Bruce and his colleagues collected background information on more than 3,000 parolees related to criminal history, personal characteristics, and prison experience and found 21 factors that differentiated those offenders who successfully completed parole from those who failed to complete parole. While some of the factors measured on the Bruce and colleagues’ Burgess Scale seem out of date (e.g., criminal type, social type, statement of trial judge and prosecutor), other measures appear on many validated risk assessment instruments used today (e.g., previous record, where offender resides, marital status).
The Burgess Scale was one of the earliest examples of an actuarial-based risk assessment tool, but along with other later-developed tools, such as the Salient Factor Score, these types of instruments have several limitations. One of the primary limitations, for instance, is that they are almost all based on static predictors: items that cannot change (e.g., criminal history) or cannot be easily influenced (e.g., age or months served prior to parole). To the extent measures other than criminal history are examined, such items capture only historical data. That is, second-generation tools may examine factors such as whether the offender had a substance abuse problem or whether the offender associated with other criminal associates, but these factors can only be examined from a historical perspective rather than a current perspective.
Third-generation instruments represented a major advancement in risk assessment since they incorporated both static and dynamic risk factors. Recall that dynamic risk factors are those items related to criminal behavior that can be changed, such as getting a job, stopping illegal drug use, or spending time with others with criminal behavior issues. Importantly, third-generation instruments such as the Level of Service Inventory–Revised can be used for both initial assessment and reassessment to measure offender behavior change over time.
Finally, the latest generation of risk assessment tools—fourth-generation instruments—is comprehensive in that they include both static and dynamic risk factors as well as take into consideration factors or barriers to treatment and intervention (e.g., lack of motivation, low intelligence, mental health issues, physical disabilities). While these are not risk factors per se, they are related to case planning and can help practitioners properly match offenders to treatment and intervention strategies so that their likelihood of success can be maximized. In addition, fourth-generation tools, such as the Level of Service/Case Management Inventory, the Ohio Risk Assessment System, and the Correctional Offender Management Profile for Alternative Sanctions, incorporate case management strategies to more clearly link risk assessment results with intervention and treatment strategies.
Types of Assessment Tools
There are three unique types of risk assessment tools. Screening instruments are designed to allow corrections professionals to determine outcome (e.g., recidivism) for a large offender population (usually at intake) and, subsequently, sort those offenders into risk categories relatively quickly. A screening instrument, for example, may be used to screen out low-risk offenders so that a more comprehensive risk assessment may be completed on those offenders who are at moderate risk to high risk. Notably, screening tools can be completed by a file review and/or a brief face-to-face interview, as they are usually composed of only static items (e.g., criminal history). The tools, however, have little utility beyond identifying levels of risk due to their limited number of dynamic items.
In addition to screening tools, comprehensive risk/need assessments are designed to gather more complete data to assist in determining levels of risk. Risk/need instruments typically take longer to administer, require training, and can be used to identify targets for treatment. These instruments include a range of dynamic factors that can be used to reassess the offender. Furthermore, composite risk assessments can be used to develop case plans to ensure that programming targets the criminogenic needs of the offender.
Once a comprehensive risk assessment is completed, specialized assessments can be used to assess specific domains (e.g., substance abuse) or special populations (i.e., sex offenders, mentally ill offenders, psychopaths). Specialized assessments, in particular, may be completed on offenders who are flagged as risky on a composite risk/needs tool. These instruments usually require additional training on the part of those who administer the assessments, while some tools require special licensures to conduct.
In many instances, jurisdictions adopt all three types of assessments. A screening instrument might be used at pretrial or to assess offenders at intake to determine who needs further assessment. For those offenders who continue deeper into the system, a more comprehensive assessment tool could be used. Once assessed on a composite risk tool, a specialized assessment may be prescribed on an as-needed basis. Following this approach can increase efficiency since not all offenders will be thoroughly assessed, but those offenders who appear to pose the greatest risk to reoffend will be examined much more closely.
Widespread Use of Risk Assessment Tools
As of 2017, risk and needs assessments are used at nearly all points in the criminal justice system across the United States, including pretrial, sentencing, probation and parole, and prison intake and reentry. Although evidence exists to show that risk assessment tools are widely used across the country, variability exists in terms of which tools are used and how they are implemented within a specific agency or department. For example, some states may adopt and implement one standardized instrument that is used across each sector of the criminal justice system and throughout the entire state. Other states, however, may take a less systematic approach and allow different criminal justice entities to choose their own method for assessing and classifying offenders.
Beyond the support for use of validated risk assessment instruments given at the national, state, and local levels, evidence exists that legislative action and statutory changes have taken place in response to risk assessment instruments. For example, in Indiana, the Supreme Court held that judges may consider risk assessment tools to supplement other evidence when making decisions about whether to sentence offenders to probation, alternative treatment facilities or programs, and other corresponding sentencing matters. Additionally, some grant administrators and criminal justice policy development agencies, such as the Bureau of Justice Assistance, oftentimes require or strongly suggest that agencies responding to funding solicitations use a validated risk assessment instrument to ensure they are adhering to evidence-based practices.
Risk Level Classification Issues
Overall, research suggests that the more commonly used risk assessment instruments used in the field can, with a moderate level of accuracy, predict who is at risk of recidivism. In addition, research suggests that when it comes to the predictive validity of the tools (i.e., whether the tools are measuring what they are supposed to be measuring), no one instrument is better or more superior than another at predicting recidivism. To this end, it is not necessarily about what instrument is being used (as long as it is an appropriate, validated tool), but about how the instrument is being used.
There are a number of important considerations with any risk/need assessment process that need to be considered. For example, how easy is the instrument to use and score? How long does it take to complete? How much training is involved? Is an interview involved, and if so, how is the information verified? How much does it cost? Beyond some of these more practical issues, there are also a number of methodological issues. Of vital importance, for example, is ensuring the instrument is both reliable and valid. Reliability refers to the consistency of the instrument. For instance, if two individuals were to conduct an assessment of the same offender, how similar would they score him or her? This is referred to as interrater reliability and can be problematic with more dynamic instruments. The second important consideration is validity or the accuracy of the instrument in predicting what it is we want the tool to predict (e.g., recidivism). Both reliability and validity are important because an accurate tool is useless if no one can agree on the score. Likewise, there might be agreement as to the score, but if it does not predict what is hoped for, then it does not lead to the goals of reducing offender recidivism and enhancing public safety. As a general rule, most good instruments are about 70% accurate. This is determined by validation studies in which the correlation between the total score and the outcome of interest (usually recidivism) is examined.
Considerable issues may also arise when agencies or departments fail to use risk assessment instruments properly. Improper use of a risk assessment tool, for instance, may lead to highrisk offenders being released from community supervision or prison before they receive an appropriate level or amount of treatment to reduce their risk. Conversely, if low-risk offenders are given too much intervention, such as being placed on intensive supervision or being sentenced to a correctional institution for an inappropriate length of time, they use valuable, oftentimes scarce resources that could be directed toward higher risk offenders. Perhaps even more important, treating low-risk clients and/or mixing them with high-risk clients has the potential to increase recidivism rates. Low-risk offenders’ rates of reoffending may increase because placing them on higher levels of supervision and providing treatment often means taking away what makes them low risk to begin with (e.g., prosocial peers, supportive family, stable employment). In a similar manner, research suggests that when low-risk offenders are mixed with highrisk offenders in treatment and supervision settings, lower risk offenders oftentimes learn things from higher risk offenders that they may not have known before (e.g., where to buy or sell drugs).
Issues related to the use of third- and fourthgeneration risk tools may also arise since these instruments include a variety of dynamic risk factors. That is, the human element or staff biases can oftentimes get in the way of any scoring or decision-making processes underlying the tool because they require the assessor to make subjective judgments. In this vein, what makes modern-day risk assessments advantageous (i.e., their dynamic measurement of risk factors and the quality and range of information collected) can also make them disadvantageous because they may require more specialized knowledge or training on how to use the tools, and/or a deeper understanding of the assessment process, in general.
Finally, another major concern is related to the use of a risk assessment tool in one jurisdiction when it has been constructed and validated in another. Undoubtedly, the development of a standardized risk instrument is often costly and the option that very few criminal justice agencies have the time or financial resources to construct. In this way, many adopt an off the shelf tool that may be empirically derived but has been constructed for a different target population. While some may argue that using a tool that has been validated on other samples is better than using no instrument at all (or a nonvalidated tool), agencies run the risk of using a tool that may not be predictive for their population. In addition, they may be placing offenders in the wrong risk groups, as cutoff scores for the tool may not be appropriate. Given that it is unlikely for a single instrument to have universal applicability across various offender populations, there is a clear need to norm (i.e., determine what is typical in a given population) and validate risk instruments and their corresponding risk scales for the population for which it is being used.
Final Thoughts
There are a number of reasons that the effective classification and assessment of offenders is important in corrections. First, results of the instrument can be used by practitioners and other corrections officials to make important decisions, such as who should be released on parole. Second, it helps reduce bias by eliminating extralegal factors from consideration, such as race and gender. Third, it enhances public safety by allowing correctional professionals to identify higher risk offenders. Fourth, it helps corrections professionals manage offenders in a more efficient manner. That is, it aids in the development of offender caseloads and workloads that are based around risk and needs so that scarce resources can be reserved for those who need them most. Fifth, the use of risk assessment instruments helps staff justify their decisions and diminish legal challenges when they arise. Ultimately, using validated instruments that are backed by science and incorporate a structured process to make decisions is more defensible than relying on gut feelings. The sixth, and perhaps the most important reason
References:
- Andrews, D. A., & Bonta, J. (2010). The psychology of criminal conduct (5th ed.). Cincinnati, OH: Anderson.
- 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., Bonta, J., & Wormith J. S. (2006). The recent past and near future of risk and/or need assessment. Crime and Delinquency, 52, 7–27.
- Bonta, J. (2002). Offender risk assessment: Guidelines for selection and use. Criminal Justice and Behavior, 29, 355–379.
- Bonta, J., & Wormith, S. J. (2008). Risk and need assessment. In G. McIvor & P. Raynor (Eds.), Developments in social work with offenders (pp. 131–152). London, UK: Jessica Kingsley.
- Burgess, E. W. (1928). Factors determining success or failure on parole. In A. A. Bruce, A. J. Harno, E. W. Burgess, & J. Landesco (Eds.), The workings of the indeterminate-sentence law and parole system in Illinois (pp. 221–234). Springfield, IL: State Board of Parole.
- Flores, A. W., Russell, A. J., Latessa, E. J., & Travis, L. F. (2005). Evidence of professionalism or quackery: Measuring practitioner awareness of risk/need factors and effective treatment strategies. Federal Probation, 69(2), 9–14.
- Jones, D. A., Johnson, S., Latessa, E. J., & Travis, L. F. (1999). Case classification in community corrections: Preliminary findings from a national survey. In Topics in community corrections (pp. 4–10). Washington, DC: National Institute of Corrections, U.S. Department of Justice.
- Jones, P. (1996). Risk prediction in criminal justice. In A. T. Harland (Ed.), Choosing correctional interventions that work: Defining the demand and evaluating the supply (pp. 33–68). Thousand Oaks, CA: Sage.
- Latessa, E. J. (2003–2004, Winter). Best practices of classification and assessment. Journal of Community Corrections, X11(2), 4–6, 27–30.
- Latessa, E. J., & Lovins, B. (2010). The role of offender risk assessment: A policy maker guide. Victims and Offenders, 5(3), 203–219.
- Latessa, E. J., & Lowenkamp, C. T. (2005a). The role of offender risk assessment tools and how to select them. For the Record (Fourth Quarter), 18–20.
- Latessa, E. J., & Lowenkamp, C. T. (2005b). What are criminogenic needs and why are they important. For the Record (Fourth Quarter), 15–16.
- Lowenkamp, C. T., Latessa, E. J., & Holsinger, A. (2006). The risk principle in action: What we have learned from 13,676 offenders and 97 correctional programs. Crime and Delinquency, 52, 1–17.
- Travis, L. (1990). Risk classification in probation and parole: Some alternative suggestions. International Journal of Offender Therapy & Comparative Criminology, 30–31, 125–139.
- Vera Institute of Justice Center on Sentencing and Corrections. (2011). Risk and needs assessment: Memorandum to the Delaware Justice Reinvestment Task Force, 9–12.
- Wright, K., Clear, T., & Dickson, P. (1984). Universal application of probation risk assessment instruments: A critique. Criminology, 22, 113–134.