DescriptionThe concept of violence against women in politics, VAWIP, grew out of women’s experiences with violence in the public sphere and around the world. This project expands on existing empirical and conceptual understandings of VAWIP by focusing primarily on (i) online violence and (ii) multiple forms of discrimination. I argue that to more effectively understand VAWIP online and against multiply-marginalized women, we need to develop new conceptualizations that center the unique considerations of these spaces and groups. Using a multi-method approach that includes grounded theory from interviews and qualitative analysis of Twitter data, I develop a typology of online VAWIP that includes target, form, and response. I then apply this typology comparatively to understand online VAWIP in Mexico and the United Kingdom. I use supervised machine learning and qualitative, thematic analysis to examine an original dataset of 1.3 million tweets that include the usernames of 77 national-level women politicians. I find that women in the U.K. receive a higher proportion and higher number of abusive posts than Mexican women politicians under analysis during the same period. I also find that online and offline violence are related and are not experienced as separate phenomena by political women. Discourses of online violence are correlated with offline forms of violence, including societal discriminations. Furthermore, online VAWIP is intersectional, in target and in form. Women politicians are more likely to be targeted with violence that affects or is related to an identity they hold. However, this finding is not identical or consistent across Mexico and the United Kingdom. Some, but not all, multi-marginalized women are targeted with greater online violence than their colleagues. Some salient discriminations are pervasive in online VAWIP posts, while others are absent or limited. Despite the complexity of these results, I find that overall, perpetrators of online violence have a broader toolbox of discriminatory rhetoric to draw from when targeting women with multiple, marginalized identities. These tools, though they vary across individuals and between contexts, serve to render women invisible and incompetent in politics and in the online space. By focusing exclusively, or even primarily, on sexism, research is discounting the forms and types of VAWIP experienced by multiply-marginalized women and viewed by multiply-marginalized audiences. In seeking to better understand the forms of violence used to delegitimize women’s access to the public sphere, patterns and shared experiences are illuminative, but so too are differences.