GDGrant Duwe

Grant Duwe

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Grant Duwe has a total of 31 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Back to the future for the Prison Rape Elimination Act (PREA): The development and validation of an actuarial risk assessment instrument for sexual violence in prison

    by Valerie A. Clark, Grant Duwe, Susan McNeeley

    Research Summary: This study evaluated the predictive validity of Minnesota's professional judgment‐based Prison Rape Elimination Act (PREA) screening tool and developed a gender‐specific actuarial instrument to assess risk for sexual victimization and perpetration in prison. Using data from 22,238 admissions to Minnesota's prison system between 2021 and 2025, we examined official reports of sexual abuse. The professional judgement‐based PREA tool demonstrated weak predictive discrimination. In contrast, the actuarial models showed substantially stronger performance, with area under the curve (AUC) values ranging from 0.74 to 0.79 for women and from 0.71 to 0.73 for men. These findings indicate that actuarial, gender‐specific tools outperform professional judgment‐based instruments and highlight the importance of tailoring risk assessments to population‐specific characteristics. Policy Implications: The results underscore the need for corrections agencies to adopt validated actuarial risk assessment instruments for PREA compliance. Current screening tools, developed primarily through professional judgment, may misclassify risk and fail to prevent sexual violence effectively. Implementing actuarial, gender‐specific tools can improve housing, programming, and supervision decisions, thereby enhancing institutional safety. Agencies should consider leveraging existing screening data to develop customized instruments and explore multistate collaborations to create tools with broader generalizability. Additionally, reassessment protocols for individuals incarcerated longer than 1 year could further improve predictive accuracy. These steps would advance PREA's goals of preventing sexual abuse and promoting safer correctional environments.

  • What Makes Prisons Safer? Examining the Relationship Between Misconduct and Involvement in Structured Activities

    by Grant Duwe, Susan McNeeley, Valerie Clark, Michael Palmieri

    Using both survey and administrative data, we examined the relationship between prison misconduct and involvement in structured activities, which include programming, work assignments, education classes and homework, and religious services, as well as unstructured activities such as physical recreation, prayer, and watching television. The results showed that greater involvement in structured activities significantly predicted both general and violent prison misconduct. Participation in structured activities not only decreased the risk of time to first general and violent misconduct, but it also led to significantly fewer misconducts overall. Additional analyses revealed that greater involvement in work assignments significantly reduced general misconduct, while more time spent on homework significantly decreased violent misconduct. The overall measure for unstructured activities did not significantly affect misconduct, although time spent in prayer reduced violent misconduct, while the amount of time writing “kites” (messages directed to staff) increased both general and violent misconduct.


  • Perspectives on Correctional Climate: Comparing Attitudes About Safety, Wellness, and Life Outlook Among Prison Staff and Incarcerated People

    by Daniel S. Lawrence, Bryce E. Peterson, James R. Coldren, Grant Duwe, Brittany C. Cunningham

    Correctional climate—the perceptions staff and incarcerated people hold about their institution—can influence well-being for everyone living and working in prison. This study explores differences in perceptions of safety, quality of sleep, psychological and emotional wellness, physical health, and overall life outlook among 1,148 security staff, non-security staff, and incarcerated individuals across five Minnesota Department of Corrections facilities. Results indicate that both staff groups report significantly greater safety and sleep quality outcomes than incarcerated people, while security staff also reported better emotional wellness than the incarcerated population. No differences emerged across groups regarding perceived psychological wellness, physical health, or life outlook. These findings highlight both shared experiences and key disparities within correctional environments, underscoring opportunities to improve conditions, support workforce well-being, and advance institutional policies that benefit all people in prison settings.


  • The Effects of Physical Victimization in Prison on Perceptions of Personal Safety and Culture of Violence

    by Grant Duwe, Susan McNeeley, Valerie Clark, Michael Palmieri

    The effects of victimization on residents in prison is an understudied area within the corrections literature. To help fill this gap, we used administrative and survey data on a sample of about 1,600 residents in Minnesota's prison system to examine the relationship between being the victim of a physical attack and perceptions of safety and a culture of violence. The results showed that residents were more likely to report being attacked in prison within the previous year when they were in prison for a sex offense, were younger in age, and had more prior misconducts. Victimization was negatively associated with perceptions of personal safety while, at the same time, positively related to the extent to which residents adopted a culture of violence. Perceptions of decreased safety were also found for women, Black residents, individuals in higher custody levels, and those who reported greater difficulty sleeping. Endorsing a culture of violence was significantly more likely for men, younger residents, individuals with more mental health concerns, and those reporting more problems relating to hunger and sleep.

  • Advancing the Prediction of Prison Misconduct: Revalidating the Minnesota Severe and Frequent Estimate for Discipline (MnSafeD)

    by Grant Duwe

    In June 2024, the Minnesota Department of Corrections implemented the Minnesota Severe and Frequent Estimate for Discipline (MnSafeD), a fully-automated, gender-specific classification assessment that was designed to predict prison misconduct. Customized to Minnesota’s prison population, the MnSafeD was created to identify the relatively small group of residents disproportionately responsible for much of the misconducts and violence that take place in prison. This study revalidated the MnSafeD on the first 8,631 residents who were scored on the instrument. The results showed the MnSafeD performed very well overall, achieving an area under the curve (AUC) of 0.84. The instrument’s predictive accuracy was better for the men and for the post-intake reassessments (0.86). The results revealed there were no predictive disparities by race/ethnicity.


  • Validating Responsivity Assessments for Correctional Populations: Evaluating the Association With Program Participation, Dosage, and Completion

    by Grant Duwe, Valerie Clark

    While specific responsivity factors are believed to either inhibit or facilitate successful involvement in programming, relatively little attention has been given to the validation of responsivity assessments used for correctional populations. Based on a sample of nearly 2,100 individuals confined in Minnesota prisons, this study examines the relationship between a needs and responsivity assessment system and multiple measures of program participation, dosage, and completion. The responsivity domains—childhood trauma, mental health, religiosity, motivation and learning style—were significantly associated with at least one measure of programming. All but two of the needs domains also had a significant association with programming involvement. We discuss the implications of the findings for correctional research and practice.


  • When Prison Becomes the Devil’s Workshop: The Association Between Idleness and Post-Release Employment, Recidivism, and Mortality

    by Grant Duwe, Valerie Clark, Susan McNeeley

    The existing literature has yet to examine the system-wide effects of idleness and participation in prison-based work and programing on multiple post-release outcomes. Using a sample of more than 77,000 releases from Minnesota prisons between 2010 and 2021, we examined the relationship between participation in prison labor and programing on post-release employment, recidivism, and mortality. The findings suggest that how people spend their time in prison significantly affects their chances of finding a job, their likelihood of recidivism, and how long they live following release from prison. As involvement in work and/or programing increased, the people released from prison were more likely to find employment while also having a reduced risk for recidivism and mortality.


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

    by Grant Duwe, Valerie Clark, Susan McNeeley

    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.


  • The Development and Validation of the Adult Dynamic Validated Instrument for Sex Offense Recidivism (ADVISOR)

    by Grant Duwe, KiDeuk Kim, Ashlin Oglesby-Neal

    This study describes the development and validation of the Adult Dynamic Validated Instrument for Sex Offense Recidivism (ADVISOR), which is designed to predict two types of recidivism—repeat sexual offending and failure to register (FTR)—specific to people convicted of sex offenses. Developed and internally validated on more than 40,000 men and women convicted of sex offenses who had been released from prison, the ADVISOR had adequate predictive performance, achieving an AUC of 0.71 for sex offense recidivism and 0.73 for FTR recidivism. The ADVISOR performed especially well for females for sex offense recidivism, obtaining an AUC of 0.78 in the internal validation and an AUC of 0.82 in the external validation.


  • New Insights for “What Works”? Religiosity and the Risk-Needs-Responsivity Model

    by Grant Duwe, Byron R. Johnson

    The existing literature has devoted relatively little attention to the responsivity principle and, in particular, religiosity as a type of specific responsivity. Using a sample of nearly 2,000 people incarcerated in Minnesota’s prison system, this study examined the relationship between religiosity and measures of risk, needs, and responsivity. The results showed that religiosity was positively associated with childhood trauma, pro-social identities, and perceived social support, while it was negatively associated with criminal thinking and employment needs. Findings further revealed that religiosity was associated with less prison misconduct and reduced recidivism risk, but only for people with positive self-identities, according to Paternoster and Bushway’s Fear of Self scale, and lower levels of perceived social support. We discuss the implications of these results for theory and practice.