Evaluating Bias, Shrinkage and the Home-Field Advantage: Results from a Revalidation of the MnSTARR 2.0

Duwe, Grant

doi:10.1080/23774657.2021.2011802

Cited by 8 publications

(9 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Second, more incarcerated individuals would have to participate in evidence-based programing, which would reduce their MnSTARR risk probabilities (Duwe, 2014(Duwe, , 2021Duwe & Rocque, 2019). In doing so, more individuals would be eligible for early release as participation in programing not only reduces MnSTARR scores, but also reduces the likelihood of recidivism (e.g., Duwe, 2013).…”

Section: Discussionmentioning

confidence: 99%

“…Today's prison administrators have the benefit of valid and reliable risk assessment tools that can predict an individual's risk of re-offense with a relatively high level of accuracy. These tools include, for example, the Level of Service/Case Management Inventory (LS/CMI; Andrews et al, 2000), the Correctional Offender Management Profiling for Alternative Sanctions assessment (COMPAS; Brennan et al, 2009), and the Minnesota Screening Tool Assessing Recidivism Risk (MnSTARR; Duwe, 2014Duwe, , 2021Duwe & Rocque, 2019).…”

Section: Prior Research On Improvised Early Release Programsmentioning

confidence: 99%

“…In addition to producing four categories of recidivism risk, the MnSTARR also produces probability scores for specific types of recidivism, including non-sexual violent recidivism, non-violent recidivism, and felony recidivism. This tool was customized to the MnDOC’s correctional population, and its most recent version (the MnSTARR 2.0) has undergone multiple validation studies (Duwe, 2021; Duwe & Rocque, 2019). Moreover, it is based on several items beyond criminal history, including but not limited to participation in evidence-based correctional programing, whether the individual received visits while incarcerated, marital status, and risk of self-injury.…”

Section: Prior Research On Improvised Early Release Programsmentioning

confidence: 99%

See 2 more Smart Citations

The Benefits of Hindsight and Modern Risk Assessment Tools: An Evaluation of Minnesota’s COVID-19 Early Release Program

Clark

2022

Crime & Delinquency

View full text Add to dashboard Cite

Given the devastating effects of COVID-19 on congregate living settings, many correctional agencies have been forced to expedite the release of incarcerated persons beginning in 2020. This research evaluated the recidivism outcomes of 158 individuals released early from prison in response to the pandemic compared to a similar cohort of standard releases. Cox regression analyses revealed that individuals released early were no more or less likely to be re-arrested or returned to prison for a supervision violation compared to the control sample, but they were significantly less likely to be convicted of a new offense. This research demonstrated that carefully selected low-risk individuals can be released early from prison without compromising public safety.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Prior Research On Improvised Early Release Programsmentioning

confidence: 99%

Section: Prior Research On Improvised Early Release Programsmentioning

confidence: 99%

See 1 more Smart Citation

The Benefits of Hindsight and Modern Risk Assessment Tools: An Evaluation of Minnesota’s COVID-19 Early Release Program

Clark

2022

Crime & Delinquency

View full text Add to dashboard Cite

show abstract

“…we also included the number of prior discipline convictions between the most recent prison admission date and the survey administration date. In addition to custody level, we included classification score, which measures risk of misconduct, as well as MnSTARR (Minnesota Screening Tool Assessing Recidivism Risk) risk level, which assesses risk for recidivism (Duwe, 2021). 3 Combined, these instruments contain items that have been found to accurately assess risk for misconduct such as criminal history, offense type, sentence length, and mental health concerns.…”

Section: Independent Variablesmentioning

confidence: 99%

Does Criminal Thinking Predict Prison Misconduct? An Evaluation of TCU’s Criminal Thinking Scales

Duwe

Clark

McNeeley

2023

Criminal Justice and Behavior

Self Cite

View full text Add to dashboard Cite

To date, only one published study has tested the predictive validity of the Texas Christian University–Criminal Thinking Scales (TCU-CTS), and no studies have tested whether these scales are predictive of prison misconduct. Using a sample of more than 2,000 people incarcerated in Minnesota’s prison system, this study examined the predictive validity of the TCU-CTS with multiple measures of prison misconduct. The results showed the overall TCU-CTS score significantly predicted misconduct, although the strength of this association was relatively modest (AUC = 0.62). Among the six scales on the TCU-CTS, Cold-Heartedness significantly predicted both measures of misconduct, whereas Power Orientation and Criminal Rationalization were each significantly associated with one misconduct measure.

show abstract

“…Before its development, MnDOC used the Level of Service/Case Management Inventory (LSI-R) to assess risk and need. When comparing the performance between the two tools, the MnSTARR outperformed the LSI-R, leading MnDOC to switch tools (Duwe, 2021). The Modified Positive Achievement Change Tool (M-PACT), developed for Washington State (Hamilton et al, 2019), has also proven to be effective in improving performance when tailored for the local jurisdiction.…”

Section: Key Takeaways From Focus Groupsmentioning

confidence: 99%

Imperfect Tools: A Research Note on Developing, Applying, and Increasing Understanding of Criminal Justice Risk Assessments

Applegarth

Lewis

Rief

2023

Criminal Justice Policy Review

View full text Add to dashboard Cite

This article shares considerations for designing, implementing, and understanding risk assessments used to reduce recidivism of people under community supervision. These insights are gleaned from 27 data scientists who participated in focus groups during the National Institute of Justice’s Recidivism Challenge Winners Symposium. Analyses revealed three primary themes: design considerations, implementation, and increasing awareness and understanding of risk assessments. Critical aspects of the design phase include validating the tool, incorporating field data that account for real-time changes, and adopting strategies to address false positives/negatives and the model’s complexity. Upon the tool’s development, practitioners are recommended to devise an implementation plan, balance attention to risk with client-focused needs, and exercise modest discretion while considering algorithmic results. Recognizing the value predictive instruments bring to decision-making and identifying their limitations is needed to increase understanding for all stakeholders. Collaboration and dialogue between tool developers and practitioners are crucial at every stage.

show abstract

Evaluating Bias, Shrinkage and the Home-Field Advantage: Results from a Revalidation of the MnSTARR 2.0

Cited by 8 publications

References 40 publications

The Benefits of Hindsight and Modern Risk Assessment Tools: An Evaluation of Minnesota’s COVID-19 Early Release Program

The Benefits of Hindsight and Modern Risk Assessment Tools: An Evaluation of Minnesota’s COVID-19 Early Release Program

Does Criminal Thinking Predict Prison Misconduct? An Evaluation of TCU’s Criminal Thinking Scales

Imperfect Tools: A Research Note on Developing, Applying, and Increasing Understanding of Criminal Justice Risk Assessments

Contact Info

Product

Resources

About