Fig. 4: Self-promotion by women is associated with fewer gains in tweet mentions than self-promotion by men.
From: The gender gap in scholarly self-promotion on social media
This plot shows the marginal effect of self-promotion (vs. no self-promotion) for both genders by year, based on a negative binomial regression fitted to 618,742 (paper, author) pairs where the author is active on Twitter at the paper’s publication time (Table 4). The model additionally controls for author’s follower count (log-scaled) and adds an interaction term between gender and self-promotion status (binary). Predictions are based on setting all control variables at their median values in the data. Error bars indicate 95% bootstrapped confidence intervals.
