Predicting Psychopathology in Jewish Ultra-Orthodox IPV Survivors: A Machine Learning Approach

Abstract

The nature of the abuse, cultural and religious values, trauma-related cognitions, and recovery actions are considered factors that shape intimate partner violence (IPV) survivors’ recovery and pathology. However, less is known about their specific impact on women’s psychopathology and wellbeing. Concomitantly, there is scant information about IPV survivors from collectivistic societies such as the Israeli Jewish Ultra-orthodox (JUO) community. The present study was designed to identify predictors of post-traumatic stress (PTSD) symptoms and wellbeing in women from the JUO community who have experienced IPV. Women (N = 261) provided information about their demographics, the nature of the violence, attitudes with respect to cultural and religious norms that normalize violence, trauma-related cognitions, the coping constructs of disengagement, faith, and engaging in help-seeking and recovery actions, and the PTSD symptoms that affect their wellbeing. A Random Forest machine learning (ML) algorithm was used to identify the strongest predictors of psychopathology and wellbeing. Regression trees were developed to identify individuals at greater risk of PTSD symptoms but also of greater wellbeing. Higher self-stigma and the perception of an unsafe world were associated with PTSD symptoms, whereas lower self-stigma, greater faith, and engagement in steps toward recovery were associated with greater wellbeing. These findings highlight the importance of treating women’s self-stigma and perceptions of an unsafe world while also encouraging faith and active engagement in recovery to promote survivors’ wellbeing and lessen their PTSD symptoms.