Election Forecasts with Twitter - How 140 Characters Reflect the Political Landscape

Abstract

This study investigates whether microblogging messages on Twitter validly mirror the political landscape offline and can be used to predict election results. In the context of the 2009 German federal election, we conducted a sentiment analysis of over 100,000 messages containing a reference to either a political party or a politician. Our results show that Twitter is used extensively for political deliberation and that the mere number of party mentions accurately reflects the election result. The tweets' sentiment (e.g., positive and negative emotions associated with a politician) corresponds closely to voters' political preferences. In addition, party sentiment profiles reflect the similarity of political positions between parties. We derive suggestions for further research and discuss the use of microblogging services to aggregate dispersed information.

Keywords: Twitter, microblogging, information market, prediction markets, election forecasts, politics, elections, sentiment analysis