The Real-World Application of the Risk Principle 1: Is It Possible in the Field of Probation?

ALTHOUGH SEVERAL CRIMINAL justice theories exist as a roadmap to effective supervision, the Risk, Needs, and Responsivity (RNR) model dominates the scholarly literature. A quick search in the Criminal Justice Abstracts Database reveals 140 peer-reviewed publications referencing the RNR model since 2000. As Andrews and Bonta (2007) note, the RNR model has been used, with increasing success, in North American and around the world. The authors further comment on the robustness of the model, but observe (2007:15) that "The greatest challenge is transferring the RNR model into 'real world' settings." Research on correctional services and the risk principle certainly supports this contention.In a review of 38 correctional programs, researchers found only 1 program that met the criteria for varying programming intensity and duration by risk (Lowenkamp, 2004). Similarly, Lovins (2012) found that 36 out of 134 correctional treatment programs reviewed were varying program duration by risk. Finally, Lowenkamp, Pealer, Smith, and Latessa (2004) found that approximately 33 percent of supervision-based programs in Ohio were targeting high-risk offenders or varying program duration by risk or program intensity by risk. Further, only four programs were meeting all three of these criteria. Echoing Andrews and Bonta (2007), it does seem that translating the RNR model in real-world settings is a challenge and has been quite elusive.A recent attempt to bring the RNR model to probation supervision was presented in a monograph titled Dosage Probation (Center for Effective Public Policy, 2014). This model drew upon the extant research on the RNR model as well as the emerging (but limited) research on dosage. One of the aspects of this model focused on setting correctional service thresholds by risk level. That is, high-risk offenders would receive 300 hours of treatment, which would require longer periods of supervision compared to moderate-risk offenders, who would receive 100 hours of treatment, thereby requiring shorter periods of supervision. This in effect would lead to varying the duration and intensity of services by risk level for offenders placed on supervision. A demonstration project of this model was developed and carried out. The authors of the evaluation found that the model had no effect on offender outcomes; however, the model (either due to complexity or some other reason) was not fully implemented (Lowenkamp, Holsinger, & Bechtel, 2016). Although it may intuitively seem easy, the struggle to properly implement the risk principle lies in the details of implementation and the availability of resources.In the federal probation system, there has been a concerted effort to align supervision practices with the RNR model since 2009. This article examines the systems effectiveness in implementing the Risk Principle. Additionally this paper examines the adoption of a violence assessment and how this can further refine the use of this Risk Principle.History of the Risk Principle in the Federal SystemThe federal probation system has a long history of using various risk prediction assessments. Individual districts used various forms of standardized risk assessment instruments throughout its history. In 1982, the Federal Judicial Center (FJC) collected survey data from across the federal system and learned that over two dozen risk prediction tools were in use (Vance, 2011). Soon after, the FJC created the RPS-80 for the entire federal probation system. The RPS-80 evolved into the RPI, which was the systems main risk assessment through 2010 (Vance, 2011).In 2008, the Administrative Office of the United States Courts (AOUSC) embarked on the creation of a fourth-generation risk assessment instrument for federal probation. In 2010, implementation of the Post Conviction Risk Assessment (PCRA) began. As Vance (2011) recounts, over the course of 18 months, every chief probation officer, supervisory probation officer (over a supervision unit), and federal probation supervision officer was trained in the RNR model. …