Perceived Gender and Political Persuasion: A Social Media Field Experiment during the 2020 Democratic National Primary

Abstract

Women have less influence than men in a variety of settings for at least two reasons. First, pervasive gender stereotypes that depict women as less authoritative often cause women to be treated differently by others. Second, research indicates many women use language, tones, or mannerisms that undermine their authority. But it is not yet known how these two explanations interact with each other in real life settings. We created a social media platform to conduct a field experiment to study this issue during the 2020 Democratic Party presidential primary election in the United States. We paid Democrats to discuss their preferred candidates with each other and complete surveys before and after these interactions on our platform. We randomly assigned some respondents to appear to their partners using gendered avatars that did not match their self-reported gender identity, unbeknownst to them. We find that misrepresenting a man as a woman undermines his influence on his partner’s candidate preferences. However, misrepresenting a woman as a man does not increase her influence. We demonstrate that one reason women have lower influence is because male discussion partners are more resistant to influence. Additionally, we find evidence of gender differences in word use. These findings challenge prevailing wisdom that women simply need to “lean in” to overcome gender discrimination and suggest that solutions to improve women’s influence and leadership will require careful, simultaneous attention to both sources of gender inequality.

Social drivers and algorithmic mechanisms on digital media

“…Testing effects of current algorithms and platform design, as well as new ideas for both, requires platforms that allow experimental manipulation. Computational social scientists have begun developing bespoke social media platforms that allow to experimentally and ethically test the effect of concealing political or gender identity (UniteDem - Combs, Tierney, Alqabandi, et al, 2022;Discuss it -Combs, Tierney, Guay, et al, 2022), social engagement metrics (Fakey - Avram et al, 2020) or anti-addictive design features (Chirp - Zhang et al, 2022).…”

Participation in Higher Education Classroom Discussions: How Students’ Identities Influence Perspective Taking and Engagement

“…Female voices were often constrained due to dominant male norms (Tannen 2002). Entrenched gender stereotypes led students-male and female-to see female instructors as less authoritative and less influential in leading conversations (Combs et al 2022;Kostovicova and Paskhalis 2021). These discriminatory, gendered perceptions impacted students' participation.…”