Radicalization Pathways in the Online Context: Accounting for User Trajectories Across Extremist Spaces

Abstract

The issue of radicalization has drawn significant attention over the last decade, as high-profile mobilizations and far-right violence grow increasingly prevalent in the United States. This has been accompanied by a particular interest in radicalization in the online context, as far-right recruitment and organizing increasingly leverages the affordances of digital platforms. In this dissertation, I advance theoretical understandings of online radicalization processes by analyzing differential participation within one growing and under-examined segment of the far-right: secular male supremacist movements. Based on prior literature identifying increasing extremism in these spaces, and increasing instances of users moving between distinct but related communities, I address two specific research questions: 1) what individual and community-level factors drive differences in engagement in male supremacist communities; and 2) what individual and community-level factors predict user migration to more extreme male supremacist communities? To answer these questions, I draw on data from five male supremacist groups active across 52 forums on Reddit, drawing on longitudinal data from the Pushshift Reddit dataset to analyze three radicalization-related outcomes: rates of engagement, migration to new and more toxic communities, and disengagement. I draw on literatures on secular male supremacist ideology, differential participation and radicalization in the online context, and online engagement more generally to develop and test hypotheses regarding factors expected to predict these outcomes; I then utilize quantitative analysis, including fixed-effects negative binomial regression and discrete-time event history analysis, to identify significant predictors of differential participation and radicalization. Overall, results show that radicalization within secular male supremacist communities on Reddit is shaped primarily by group identity, certain individual features of engagement or extremity, and the qualitative components of users’ interactions within these spaces, all of which work in concern to ‘push’ or ‘pull’ users through—or anchor them within—secular male supremacist spaces. These factors then operate alongside platform-level repression efforts to influence differential participation and user trajectories into extremism. Taken together, the analysis advances scholarship by highlighting the bolstering effect of having a diverse set of extremist spaces in which users can engage, identifying important effects of toxicity on extremism, determining which features of interaction are important for driving differential participation, and implicating platform-level repression efforts in driving radicalization.