Search
Browse By Day
Browse By Time
Browse By Person
Browse By Mini-Conference
Browse By Division
Browse By Session or Event Type
Search Tips
Virtual Exhibit Hall
Personal Schedule
Sign In
X (Twitter)
Research question and hypothesis
It is well-established that male and female politicians talk about different policy topics and use different language in ways consistent with gender stereotypes. Gender differences in political communication are often attributed to differences in preferences or to division of tasks within parties. We claim that an additional reason why men and women politicians choose different topics and use different language is that citizens reward them for talking in gender stereotypical ways, and they utilize social punishment to discipline politicians who do not conform to gender norms. Using insights from cognitive psychology, we argue that citizens prefer political messages that are congruent with socially relevant categories such as gender. Because politicians are sensitive to social reinforcement, they respond to feedback by sending messages ever more aligned with their gender.
Data and Methods
To investigate our claim, we collected data from the Twitter accounts of all Spanish regional and national parliamentarians during 2018, including the text of all tweets as well as the number of replies, retweets, and likes received by every tweet. Using machine learning methods, we first confirm that male and female politicians use different words in their communication with the public. We derive an estimate of gender congruence that measures how typical is each tweet of messages sent by male or female politicians. In the next step, we regress the positive feedback received by each tweet on the gender typicality of the tweet. We include politician and day fixed effects in order to isolate the effect of the typicality of tweets from time invariant characteristics and overall period effects. In order to identify feedback effects, we regress the typicality of tweets sent by the same politicians in the next few days on the amount of feedback they receive.
Results
We find that tweets that are gender-typical receive more positive feedback than tweets that are not (typical tweets contain words typically used by women and are sent by women or contain words often used by men but are sent by men). The effects are stronger for female politicians. We also find that the feedback sent by users affects politicians’ subsequent behavior. Specifically, politicians react to receiving more positive feedback by sending more gender congruent tweets. Our results imply that citizens incentivize politicians to communicate in ways that are congruent with gender stereotypes, hence contributing to maintain them.
Aina Gallego, Institut de Barcelona d'Estudis Internacionals
Gael Le Mens, Pompeu Fabra University
Nikolas Schoell