An Examination of the Influence of Gender and Race on Dynamic Risk Assessment

Public Deposited
Resource Type
  • The need for continued research on justice involved (JI) women and individuals belonging to marginalized groups is increasingly being recognized by correctional agencies and researchers, and a body of literature examining the intersection of gender, race, and correctional outcomes has begun to develop. The goal of this dissertation was to contribute to this emerging body of knowledge by exploring the impact of individuals' gender and race on dynamic risk assessment using the Dynamic Risk Assessment for Offender Re-Entry (DRAOR). Study 1 used a sample of 3,091 racially diverse (78.2% White, 18.1% Black, and 3.7% Hispanic) JI women supervised in the community in Iowa to explore the psychometric properties and factor structure of the DRAOR. The DRAOR's original structure provided a poor fit for the data, and alternative models were identified using factor analytic techniques and exploratory structural equation modeling (ESEM). Measurement invariance testing was used to determine whether DRAOR scores could be meaningfully compared across race and across assessment occasions. Contrary to expectations, the DRAOR did not demonstrate adequate measurement invariance and implications for case management are discussed. Next, Cox regression survival analysis was used to explore whether recidivism rates differed across race in Study 2. Three recidivism outcomes, technical violations, new offences, and any return were examined. Results indicated that Black women were significantly more likely to incur technical violations than their White and Hispanic counterparts. Findings from Study 2 also indicated that survival time was significantly associated with women's level of risk, though not always in the expected direction. Study 3 used a subsample of 2,763 women and matched sample of 2,763 men to explore the predictive ability of the DRAOR across gender and race. Prediction was assessed by means of discrimination, calibration, and regression analyses. Despite considerable variability across analyses and type of recidivism examined, findings collectively indicated that DRAOR scores predicted most accurately for White men. Collectively, findings suggest that parole officers should be cautious in using the DRAOR with JI women and individuals from marginalized groups and underscores the need for additional research examining the influence of gender and race on dynamic risk assessment.

Thesis Degree Level
Thesis Degree Name
Thesis Degree Discipline
Rights Notes
  • Copyright © 2020 the author(s). Theses may be used for non-commercial research, educational, or related academic purposes only. Such uses include personal study, research, scholarship, and teaching. Theses may only be shared by linking to Carleton University Institutional Repository and no part may be used without proper attribution to the author. No part may be used for commercial purposes directly or indirectly via a for-profit platform; no adaptation or derivative works are permitted without consent from the copyright owner.
Date Created
  • 2020


In Collection: