Validating the Dynamic Risk Assessment for Offender Re-Entry (DRAOR) in a Sample of U.S. Probationers and Parolees

Abstract

The foundation for effective case management is rooted in the use of a validated risk assessment.The present study sought to validate the Dynamic Risk Assessment for Offender Re-entry (DRAOR) among a sample of probationers and parolees (n = 391) in the state of Iowa. Scores across the DRAOR domains were able to differentiate between recidivists and non-recidivists when examining technical violations and any recidivism, although were unable to differentiate between those offenders who were re-arrested and those who remained crime free. An examination of the psychometric properties of the scale suggested that the DRAOR is a valid risk assessment tool. Additionally, Stable dynamic risk factors represented the strongest predictor of technical violations, although were unable to predict rearrest. The predictive utility of the various domains (i.e. Stable, Acute, and Protective) suggested that case managers would benefit from utilizing the DRAOR in the everyday supervision of offenders. U.S. DRAOR VALIDATIONiii Acknowledgements I am extremely grateful to my supervisor, Dr. Ralph Serin, for providing me with this opportunity and for the constant support and feedback throughout the project. I am also thankful for the advice and feedback from my committee member, Dr. Shelley Brown who's comments undoubtedly improved the quality of this project. I also appreciate the knowledge and perspective shared by Dr. Diana Majury. This project would not have been possible without the cooperation of the Iowa Department of Corrections. I truly appreciate the assistance that Dr. Lettie Prell provided throughout the various stages of the project.A special thank you to Caleb Lloyd who provided unwavering support throughout the duration of the project. I am also grateful for the efforts of Mark Girvan and Eric Filleter who assisted with preparing the data.

Can a Dynamic Risk Instrument Make Short-Term Predictions in "Real Time"? Developing a Framework for Testing Proximal Assessment of Offender Recidivism Risk During Re-entry

“…between subscales and static risk (RoC*RoI) scores. At baseline, internal consistencies were similar to Cronbach's alpha estimates using DRAOR data drawn from another jurisdiction (i.e., Iowa; see Chadwick, 2014, or refer back to Table 3.2).…”

“…Several past projects have validated DRAOR as a promising tool for risk assessment, and dynamic re--assessment. These validation projects arose from the implementation of DRAOR into community supervision practice throughout Iowa (see Chadwick, 2014;Smeth, 2013) and throughout New Zealand (see Hanby, 2013;Yesberg & Polaschek, 2014;Yesberg et al, 2015). This section reviews the research listed in Table 3.1 by focusing on the psychometric properties that are necessary to justify building the statistical models Factor structure and measurement invariance.…”

“…For example, an exploratory factor analysis (EFA) loaded all risk items together, separate from protective factors, creating a two--factor scale (Chadwick, 2014). Hanby's (2013) analyses suggested that the original DRAOR structure fit her data well, but that model fit could be improved if some risk items were allowed to cross--load, or were re--located from the stable subscale to the acute subscale, or vice versa.…”

“…However, differences in training place a lower limit on the predictive ability of a risk tool rather than artificially inflating findings. For example, research with front--line staff rating the Level of Supervision Inventory--Revised (LSI--R) showed the relationship between LSI--R scores and recidivism was significant among formally (Chadwick, 2014). By contrast to the use of DRAOR in Iowa, all supervision officers who provided data for the dataset used in this dissertation were formally trained to properly rate DRAOR items.…”

Validating the Structured Dynamic Assessment Case-Management 21-item (SDAC-21) in a Sample of Incarcerated Offenders

“…In the first large scale study on the DRAOR, using a sample of 3,498 parolees in New Zealand, AUC's ranged from .67 to .72 when predicting violation behaviour (Hanby, 2013). Further, research completed in Iowa has shown that the DRAOR is valid among male general offenders (Chadwick, 2014;Serin, Gobeil, Lloyd, Chadwick, Wardrop & Hanby, 2016) and sex offenders (Smeth, 2013).…”