Year of Award


Document Type


Degree Type

Doctor of Philosophy (PhD)

Degree Name

Clinical Psychology

Department or School/College

Department of Psychology

Committee Co-chair

Gyda Swaney, Bryan Cochran

Commitee Members

Craig McFarland, Daniel Denis, Laurie Walker


Diversity, Indigenous, Native American, Recidivism, Reentry, Resilience


University of Montana


Background: Native Americans have suffered from vast losses of land, traditional ways and practices, language, and ability to pass on traditional knowledge; and those losses have led to current day health and wellness disparities, as well as small but tangibly different tribal populations. While Native Americans make up a disproportionate number of those involved in the criminal justice system, creating an overrepresentation of Native Americans incarcerated in jails and prisons relative to their overall population size, they still tend to be underrepresented in the creation of validated recidivism measures. Risk assessment tools, such as the Level of Service Index-Revised (LSI-R), that have been examined in Native American individuals do not consistently uphold predictive validity. Additionally, most evidence-based approaches for treatment also do not have a large enough sample of Native Americans who are included in research efforts to draw conclusions about the efficacy of any given approach. Research has suggested that re-validating existing measures may be a useful approach. Additionally, adding cultural factors in the assessment and treatment of Native American individuals may serve to more accurately evaluate risk for recidivism, ultimately guiding the appropriate level of intervention and treatment approach.

Method: Archival data from male and female Native American individuals participating in the Flathead Reservation Reentry Program (FRRP) between February 2016 and September 2018 were used. Participants were members of a federally recognized tribe, involved in the criminal justice system, and plan to reenter or have reentered the Flathead Reservation community from incarnation. Participants in this study included 216 Native American adults ranging in age from 18-65 years (M = 34.1 years). The sample included 133 males and 83 females, and the majority of the sample identified as enrolled CSKT (170, other tribal enrollment 46).

Results: Hierarchical logistic regression models detected statistical significance for the overall LSI-R, but only 2-3 of the subscore domains were significant upon further analyses. Cultural measures as well as intensity case management involvement were not statistically significant. Overall the models resulted in small effect sizes.

Discussion: The results of these analyses uphold the notion that the LSI-R is not a good tool for measuring recidivism risk, but other factors that were predicted to be statistically significant were also not found to be significant in the models. Suggestions and recommendations for further data collection and analysis within this population are provided.



© Copyright 2019 Desiree Lacy Fox