Measuring and Unpacking Affective Polarization on Twitter: The Role of Party and Gender in the 2018 Senate Races

Abstract

This study examines how the Twittersphere talked about candidates running for the U.S senate in the 2018 congressional elections. We classify Twitter users as Liberal or Conservative to better understand how the two groups use social media during a major national political election. Using tweet sentiment, we assess how the Twittersphere felt about in-group party versus outgroup party candidates. When we further break these findings down based on the candidates' gender, we find that male senatorial candidates were talked about more positively than female candidates. We also find that Conservatives talked more positively about female Republican candidates than they did about Republican male candidates. Female candidates of the out-group party were talked about the least favorably of all candidates. Conservative tweeters exhibit the most positive level of in-group party sentiment and the most negative level of out-group party sentiment. We therefore attribute the most intense affective polarization to this ideological group.

Does Social Media Use Promote Political Mass Polarization?

“…In addition, conservatives in the US tweeted about ingroup candidates more positively and, simultaneously, more negatively about opposing candidates than did liberals (Mentzer, Fallon, Prichard, & Yates, 2020). Moreover, more Twit ter users were found to be both positively and negatively polarized toward Hilary Clinton in comparison to Donald Trump (Grover et al, 2019), and men on Twitter appeared to voice less ingroup party support and less dislike of the out-group party than women did (Mentzer, Fallon, Prichard, & Yates, 2020). Furthermore, several studies identified homophily at work (Yardi & boyd, 2010;Gruzd & Roy, 2014), with interactions with like-min ded individuals on Twitter strengthening group identity, whereas engage ment with different-minded individuals reinforced ingroup and outgroup affiliations (Yardi & boyd, 2010).…”

Affective Polarization and Dynamics of Information Spread in Online Networks

“…Using advanced transformer-based language models we quantify emotions and toxicity of online interactions and present evidence for affective polarization: users express more negative emotions and use more toxic language in their replies to out-group members, but express more joy in replies to in-group members. Unlike previous research, which looked at how people talked about out-group members 14,18,20 , we examine how people talk to them. Importantly, we also show that expressed emotions vary with distance in the retweet network, suggesting that emotions organize social network structure.…”

“…This paper contributes to this body of work by proposing a methodology to measure affective polarization on social media and investigating the interactions between affective polarization, the structure of online networks, and dynamics of information spread on them. In contrast to existing research, which explores how people talk about out-group members 14,[18][19][20] , we focus on how they talk to them. We leverage state-of-the-art language models to measure emotions and toxicity of reply interactions and show that they demonstrate in-group favoritism-out-group animosity that is the hallmark of affective polarization 1 .…”

Questions of Communicative Change and Continuity

“…For example, US American climate-change dis believers on Twitter showed higher levels of hostility toward climate-chan ge believers than vice versa (Tyagi, Uyheng, & Carley, 2020). In addition, conservatives in the US tweeted about ingroup candidates more positively and, simultaneously, more negatively about opposing candidates than did liberals (Mentzer, Fallon, Prichard, & Yates, 2020). Moreover, more Twit ter users were found to be both positively and negatively polarized toward Hilary Clinton in comparison to Donald Trump (Grover et al, 2019), and men on Twitter appeared to voice less ingroup party support and less dislike of the out-group party than women did (Mentzer, Fallon, Prichard, & Yates, 2020).…”