Women Use More Positive Language than Men: Candidates’ strategic use of emotive language in election campaigns

         
Author Name Tiffany BARNES (University of Kentucky) / Charles CRABTREE (Dartmouth College) / MATSUO Akitaka (University of Essex) / ONO Yoshikuni (Faculty Fellow, RIETI)
Creation Date/NO. December 2022 22-E-114
Research Project Advanced Technology and Democracy: Does new technology help or hurt democracy?
Download / Links

Abstract

How do political candidates strategically use emotive language for electoral purposes? We argue that women candidates are more constrained in the strategies available to them in shielding themselves from backlash on the campaign trail. To test our theoretical expectations, we construct and use a dataset of approximately 165,000 Tweets from 2,662 candidates and responses to these Tweets that were posted during the last two UK General Elections. Our analysis of candidate Tweets finds that women candidates are more positive and less negative than their men counterparts, regardless of whether they are incumbent or challengers. Importantly, this pattern of women’s behavior is not simply reflective of socialization. Indeed, our results show that negative Tweets attract more attention (in terms of the number of replies and likes) for both men and women, but that negative Tweets from women candidates are met with more negative responses than those from men. In other words, women candidates face backlash when they engage in negative emoting. These findings suggest that, consistent with our argument, women candidates are strategically motivated to behave in gender-typical ways in election campaigns.