Talking Politics on Twitter: Gender, Elections, and Social Networks

Abstract

As campaign discussions increasingly circulate within social media, it is important to understand the characteristics of these conversations. Specifically, we ask whether well-documented patterns of gendered bias against women candidates persist in socially networked political discussions. Theorizing power dynamics as relational, we use dialectic configurations between actors as independent variables determining network measures as outcomes. Our goal is to assess relational power granted to candidates through Twitter conversations about them and whether they change depending on the gender of their opponent. Based on more than a quarter of a million tweets about 50 candidates for state-wide offices during the 2014 US elections, results suggest that when a woman opposes a man, the conversation revolves around her, but she retains a smaller portion of rhetorical share. We find that gender affects network structure—women candidates are both more central and more replied to when they run against men. Despite the potential for social media to disrupt deeply rooted gender bias, our findings suggest that the structure of networked discussions about male and female candidates still results in a differential distribution of relational power.