To what extent sentiment analysis of Twitter is able to forecast electoral results? Evidence from France, Italy and the United States

The growing diffusion of social media raises the possibility to delve into the web to explore and track the political preferences of citizens. Monitoring social media during an electoral campaign can become therefore a useful complement of traditional off-line polls. Some scholars, however, go even further than that, claiming that by analyzing social media we can produce a reliable forecast of the final result. Relying on a proper methodology for sentiment analysis remains a crucial issue in this respect. In this work, we apply the recent supervised method proposed by Hopkins and King (2010) to analyze the voting intention of Twitter-users in three different contexts: France (for the 2012 Legislative election, first round), Italy (for the two-rounds of the centre-left 2012 primaries) and the United States (for the 2012 Presidential election). Our analysis shows a remarkable ability of Twitter to forecast electoral results, in some cases improving on mass surveys results. On the contrary, more “naïve” analyses of social media provide results less satisfactory than ours.