Effects of Setting, Analyses and Probation Status, The

Corrections Compendium, Spring 2009 by Lurigio, Arthur J, Olson, David E, Snowden, Jessica

Predicting Rearrests Among Felony Probationers:

Authors' Note: This research was funded by the Cook County Adult Probation Department. The opinions expressed in this article are those of the authors only and do not represent the opinions or positions of the Cook County Adult Probation Department or the Circuit Court of Cook County.

Probation Is the most common sentence in the U.S. for adult offenders. More than four million people are currently on probation in this country, and hundreds of thousands of convicted offenders are evaluated each year to determine their eligibility for a probation sentence (Glazer and Bonczar, 2006). Probation officers are responsible for the implementation of a probation sentence (Lateesa and Allen, 2003) and also provide judges with essential information about offenders at different stages of the criminal justice process. At the presentencing stage, probation officers make critical recommendations about the dispositions of criminal cases. At the post-sentencing stage, they protect the community by enforcing the conditions of probation, monitoring offenders' progress while on community supervision and reporting violations of probation to the court. In addition, they help offenders manage their problems by brokering clinical services and other resources.

Probation officers perform many roles, and their decisions affect the legal outcomes of a substantial number of offenders. To one degree or another, most probation officers' decisions are based on an evaluation of offender risk; they decide, for example, whether to recommend probation in lieu of prison, or home confinement as a special condition of probation. Both these types of recommendations stem from probation officers' predictions about an offender's likelihood of engaging in future criminal behavior.

This article focuses on risk assessment, one of the oldest and most fundamental decision-making processes In community corrections (Holsinger, Lurigio and Latessa, 2001). The problem of predicting offender risk is paramount to decision-making at various stages of the criminal justice process and is integral to several probation decisions. The most common criterion or outcome for risk assessment In probation is rearrest; therefore, the identification of the predictors of rearrest Is the first step in the development of risk assessment tools or scales. These tools are the basis for case classification and other systems that place offenders in categories for supervision and services (Gottfredson, 1987).

In this study, the predictors of rearrest for a large sample of adult probationers in Illinois were examined by comparing the state's most populated county with the state's other counties. Various multivariate analytic techniques were used to identify the predictors of rearrest during and after the probation period. The implications of these comparative analyses for the evaluation and supervision of probationers are discussed.

Generations of Risk Assessment

When judges sentence offenders to probation, they trust that probation officers will modify their supervisory activities to lower offenders' propensity for rearrest. If the probation officers' risk-reduction efforts are unsuccessful, officers are expected to minimize the harm caused by probationers' continued criminal behavior by petitioning judges to revoke their probation, which can result In incarceration, an extension of probation or other punishments. Probation officers have always accepted this responsibility; however, during the past 30 years, the risk-assessment process has evolved. Experts in the field of community corrections have identified three generations of risk assessment: subjective or clinical judgments, actuarial or case classification approaches, and criminogenic /dynamic evaluations (Bonta, 1996).

Subjective/clinical judgments. The first generation of assessment strategies, predicated on subjective or clinical judgments and professional experiences, were used up until the early 1970s. Probation officers relied on their own sensitivity, "gut feelings," common sense and informal criteria to evaluate probationers' risks and needs and to formulate case management strategies. Clinical evaluations are drawn mostly from the Individualized views of an officer and are difficult to examine empirically (Van Voorhis, Braswell and Lester, 2000). A reliance on the stereotypical judgments of individual officers created unknown degrees of bias in decision-making and an absence of standardization and reliability in assessment and supervisory techniques (MacKenzie, 1989).

Without the guidance of standardized assessment tools, judgments are idiosyncratic (i.e., different officers use different case characteristics to render subjective decisions), susceptible to officers' preconceived (and often fallacious) notions about risk indicators, and guided by officers' personal philosophies and experiences (Wagner, 1992). The variables that officers select for risk assessment can be intuitively appealing, yet bear no statistical relationships to the predicted outcomes. Officers also are limited in their capacities to consider more than one predictor variable at a time. Risk scales, In contrast, rest on multivariate frameworks, containing several factors that can be considered simultaneously during the classification process. To replace the clinical decision-making approach to risk assessment, probation administrators began to adopt tools that were derived from statistical models.