Notes
Effects of sexism on women’s political expression online - Evidence from a preregistered repeated-measure experiment
15The invisibility of women in the political arena is an essential issue stemming from gender stereotypes and discrimination. Women overall prefer less visible forms of participation (Bode, 2017). Scholars frequently demand more (non-stereotypical) behavioral models to close such gender gaps. Yet, we argue that the role-model effect does not simply transplant in the digitally networked space. Research suggests a sexist bias against women who visibly engage in politics, resulting in incivility and harassment. The backlash against female politicians (Rheault et al., 2019) and journalists (Gardiner, 2018) sanctions women’s visible political self-disclosure to emphasize politics as a masculine space (Harrison & Munn, 2007). Relying on the theory of normative social behavior (TNSB, Rimal & Real, 2005), we assume the microcosmos of online discussions enforces social rules of political participation. For one, questioning and attacking women in online political discussions displays a descriptive social norm of the medium (Flanagin, 2017). Additionally, such public harassment suggest that women have to expect consequences when voicing their opinions (injunctive norms). Our first research question is whether there is a role-model effect of visible female politicians. Secondly, we ask whether the sexist backlash against visible women in political discussions has a silencing effect on female readers of political news online.
16We present a preregistered study (https://osf.io/s79ya/?view_only=f7cde7241b3d414dac47810c476adea1) investigating the impact of sexist comments under political news postings on social networking sites on female readers’ political engagement online (n = 500). We implemented a repeated-measure design using randomized treatment and stimuli presentation using six randomized social media posts from the leading legacy news show in Germany (Tagesschau) and the customary comment sections. Single item measurements were used for all outcomes. A power analysis was conducted as a simulation study, showing sufficient power to detect small average treatment effects under the assumption of limited effect heterogeneity at the stimulus level.
17Political discussions / gender gap / sexism / social norms / harassment / online participation