A Fundamental Cause Approach to Intimate Partner Violence in Marginalized Populations

Abstract

Intimate Partner Violence (IPV) is a serious public health concern that transcends partnership types and societies, affecting approximately one-third of women worldwide. Men in same-sex relationships likely experience violence at similar rates. This dissertation takes a fundamental cause approach to social-ecological and minority stress theories, arguing that structural forces are the root causes of violence in two target populations: women residing in low- and middle-income countries (LMIC) and Indigenous men who have sex with men (MSM) in the United States. Consisting of eight chapters, four of which are empirical studies, this body of work creates new knowledge surrounding how social scripts, environments of stigma, and contexts of inequality shape the risk for IPV in marginalized populations. Chapter 4 uses the Demographic and Health Surveys (DHS) fieldworker dataset to test for interviewer bias in the DHS Domestic Violence Module. Multilevel modeling was used to test associations between three types of IPV and socio-demographic interviewer characteristics. Previous experience as a DHS interviewer was associated with significantly lower odds (aOR: 0.67) of reporting physical IPV. In addition to highlighting a potential source of bias, this paper argues for the expanded use of the fieldworker survey to control for potential interviewer bias in the DHS. Chapter 5 studies how individual deviation from community norms shapes the risk for sexual IPV in 32 low- and middle-income countries (LMIC). Using a positive deviance approach, this analysis seeks to better understand how transcending community norms alters the risk for violence, and how this relationship changes across six structural contexts fundamental to IPV. Positive deviance is associated with both increased and decreased odds of reporting sexual IPV and the nature of these relationships vary by structural environment. Lessons from this paper may highlight pathways for future interventions to change restrictive social scripts and increase women’s social capital while avoiding an unintended increase in violence. Chapter 6 represents the first nationwide study of IPV in Indigenous MSM, a population experiencing multiple structural vulnerabilities. A 30-minute online survey consisting of instruments previously validated in LGBT or Indigenous communities was targeted to Indigenous MSM using social media algorithms. Logistic regression models were fit to calculate adjusted associations between race- and sexuality-based structural stressors, theory-derived points of resiliency, and lifetime experience of physical, sexual, and emotional IPV. Results suggest that Indigenous MSM experience high levels of IPV, and that structural stressors play a significant role as antecedents of violence Chapter 7 uses a birth cohort analysis of 25 Demographic and Health Surveys (DHS) to examine whether the socio-political environments in which a woman forms her attitudes around IPV influences its reporting, and whether this cohort effect varies by rate of decline in gender inequality. Results suggest a birth cohort effect is present in physical IPV, sexual IPV, and the justification of IPV across the 25 countries, with women reporting less IPV with each successive cohort. This effect wanes as the rate of gender inequality decline slows. Chapter 8 provides summary, conclusions, and recommendations for future work to mitigate and prevent IPV in LMIC women and Indigenous MSM.