Electronic Thesis and Dissertation Repository

Thesis Format



Master of Science


Epidemiology and Biostatistics


Bauer, Greta R.


Background: Intimate partner violence (IPV) is a recognized public health issue that can lead to poor mental and physical health outcomes. It is critical to take an intersectional approach to understanding the ways that social and interpersonal power impact IPV.

Methods: Random Forest was used to aid in social group selection when forming intersections. Multilevel analysis of individual heterogeneity and discriminatory accuracy (MAIHDA) was used to estimate the prevalence of IPV across intersections. Descriptive statistics were used to explore the context in which IPV occurred.

Results: The prevalence of IPV was greatest for cisgender women, transgender, and non-binary individuals, aged 15-24, with moderate to severe disabilities. Cisgender women, transgender, and non-binary individuals, aged 45+, with disabilities were more likely to report severe psychological consequences of the violence.

Conclusions: Public health efforts should seek to understand and address the complex structural inequities experienced by intersections at highest IPV risk.

Summary for Lay Audience

Intimate partner violence (IPV) involves behaviours which cause physical, psychological, and/or sexual harm to those within a current or former relationship. Perpetrators will use IPV to gain power and control over their partner. IPV is considered a public health issue, as it can lead to poor mental and physical health outcomes for those experiencing the violence. Certain groups are known to be at higher risk of IPV, including younger women and girls, people with disabilities, sexual minorities, and Indigenous women.

The Intersectionality Framework explains that individuals have many social identities and positions, which intersect with one another to shape one’s experiences. Since IPV is rooted in power and control, social power dynamics (e.g., privilege and oppression) influence one’s risk of IPV. An intersectional approach is necessary in understanding IPV, as allows us to understand the risk of IPV among intersection groups that have been hidden in prior research.

This study analyzed the Statistics Canada 2018 Survey of Safety in Public and Private Spaces to take an intersectional approach to describe the prevalence of IPV in Canada. We used quantitative methods that have been shown to be well-suited for taking an intersectional approach to studying health inequities. The literature and machine learning methods were applied to aid us in choosing which social groups would be used in forming the intersections for this study. We also described the context in which the violence occurred among those who experienced IPV, including consequences of the violence and help-seeking behaviours.

The social groups chosen to form intersections were sex/gender, age, and disability status. We found cisgender women, transgender, and non-binary individuals, aged 15-24 years old, with moderate to very severe disabilities experienced the greatest prevalence of IPV in the past 12 months preceding the survey. Cisgender women, transgender, and non-binary individuals aged 45 and up, with disabilities experienced the most severe psychological impact from the violence. Our findings demonstrate the importance of taking an intersectional approach to studying IPV. Intersections found to be of greatest IPV risk should be prioritized in public health prevention and intervention efforts.

Available for download on Friday, December 01, 2023

Included in

Epidemiology Commons