Public Spheres in Twitter- and Blogosphere. Evidence from the US

TL;DR

This study examines how followers of US political parties on Twitter endorse thematic versus episodic content, revealing differences between parties and the influence of election proximity on endorsement likelihood.

Contribution

It provides empirical evidence on the content preferences of political social media followers and introduces a method to analyze thematic and episodic content in tweets.

Findings

Democratic followers slightly prefer thematic messages.

Republican followers show no strong preference.

Proximity to elections increases endorsement odds.

Abstract

The political requires a forum for its deliberation and Habermas has located it in the public spheres. Originally, mass media's role was one of a facilitator of these debates. However, under the immense pressures of free market competition and mobile audiences, mass media prefers episodic over thematic news. On the opposite end of the spectrum, social media has been heralded as a new forum, a reincarnation of the ailing public spheres to further the deliberation of the political. But do the followers of political parties in social media endorse thematic or episodic content? To answer this question, I look at the most recent 3,200 tweets that were broadcast from the Republican and Democratic Twitter accounts. By employing Latent dirichlet allocation, I extract the prevailing topics of these tweets and linked websites. Generalized linear models are used to describe the relationship between episodicity, thematicity and the endorsement counts of the posts analyzed. I find that there is a stark contrast between the behavior of Democratic and Republican followers. In general, there seems to be a slight preference for thematic messages. Interestingly, the distance to an election increases the odds of a message to be endorsed.