Abstract: The 2016 presidential election illustrated the growing role that micro-blogging sites such as Twitter play in electoral politics. In this paper we report an analysis of a unique dataset that characterizes how the substantive and affective content of Tweets evolved during the course of three pivotal Republican Primary debates leading up to the 2016 Presidential election. We find that as the debates progressed Tweets provided an increasingly backward-looking account of the debates, as original content gradually gave way to retweets of the most popular earlier posts. Moreover, whereas during the debate Tweets focused on a mix of substantive topics, the Tweets that had the longest staying power after the debates were those that focused on the more sensationalist news events, often through pictures and videos. As such, a user coming to Twitter after the debate was over would have encountered a different topical and emotional landscape than one who had been following the site in real-time, one more closely resembling a tabloid than a substantive discussion forum. We explore the potential implications of the findings for the role that micro-blogging sites may have on shaping voter opinion in elections.
STAT111 - INTRODUCTORY STATISTICS
Introduction to concepts in probability. Basic statistical inference procedures of estimation, confidence intervals and hypothesis testing directed towards applications in science and medicine. The use of the JMP statistical package.