A Sex Moderated Latent Class Analysis

• In Event: The Physical and Mental Health Impacts of Incarceration

Abstract

This study presents the results of latent class analyses (LCA) and multinomial logistic regressions conducted separately on samples of self-identified females and males to identify distinct patterns of emotional distress and associated indicators within each group and assess the impact of program participation on mental health outcomes. The current study is comprised of data from the Survey of Prison Inmates, United States, 2016 (ICPSR 37692), focusing specifically on the public-use state data. The LCA results revealed multiple latent classes characterized by varying levels of anxiety, depressive symptoms, and emotional distress. The female sample yielded a 4-class model, while the male sample resulted in a 5-class model. The differences in class composition between sexes highlights the importance of sex specific mental health policies and interventions. The multinomial logistic regressions produced distinct results for each self-identified sex. For females, participation in job training programs and money management classes was associated with an increased likelihood of belonging to the Severe Emotional Distress class. Males, on the other hand, showed that anger management classes and sex offender treatment program participation displayed differential impacts across different mental health classes. Ongoing evaluation of program effectiveness and sustainability is recommended to inform policy development and improvement.

Author

• Alexa Mata, Sam Houston State University