Birds of the Same Feather Tweet Together. Bayesian Ideal Point Estimation Using Twitter Data

Abstract

Politicians and citizens increasingly engage in political conversations on social media outlets such as Twitter. In this article, I show that the structure of the social networks in which they are embedded can be a source of information about their ideological positions. Under the assumption that social networks are homophilic, I develop a Bayesian Spatial Following model that considers ideology as a latent variable, whose value can be inferred by examining which politics actors each user is following. This method allows us to estimate ideology for more actors than any existing alternative, at any point in time and across many polities. I apply this method to estimate ideal points for a large sample of both elite and mass public Twitter users in the United States and five European countries. The estimated positions of legislators and political parties replicate conventional measures of ideology. The method is also able to successfully classify individuals who state their political preferences publicly and a sample of users matched with their party registration records. To illustrate the potential contribution of these estimates, I examine the extent to which online behavior during the 2012 US presidential election campaign is clustered along ideological lines.

Can we vote with our tweet? On the perennial difficulty of election forecasting with social media

“…Nevertheless, some emerging work has suggested how it might be done. Barberá (2012) uses ideal point estimates (Poole & Rosenthal, 1985) of Twitter-using politicians as a benchmark for predicting the political alignment of Twitter users. Since ideal point estimates are easy to update for each new political season, this work provides a possible means of updating partisan forecasts without resorting to polling.…”

Who watches (and shares) what on youtube? and when?

“…Users were then labeled as left or right according to the distribution of users they followed (if any). Following had previously been shown to be a strong signal for political orientation [7,4,35,37].…”

Co-following on twitter

“…Several papers have looked at various Twitter user classification tasks, typically for (i) political orientation in the US, (ii) gender, and (iii) age [19,21,26,7,8,3]. This line of work usually involves a broad set of features, including textual content, network and activity based features, as well as a variety of classification approaches that make use of label-propagation across social links.…”

“…For both cases we look at the top discriminative co-following features from the http://WeFollow.com categories Music, Sports and News. 3 Table 2 shows some examples of the insights we can get from co-following patterns. The lifestyle correlations for the political rivalry @GOP vs. @TheDemocrats can be inspected to make intuitive sense with, e.g., @nytimes being more popular among @TheDemocrats followers.…”

“…First, there is lots of work on user classification on Twitter, e.g., [7,8,3]. Such classifiers often rely on language-specific tools such as stemming or dictionaries with special terms.…”