Partisan alignments and political polarization online: a computational approach to understanding the french and US presidential elections

With the advent of Twitter and the ability to collect large datasets from this technology, researchers have the opportunity to analyze political participation in cross-national electoral contexts. This paper capitalizes on this capability to examine political polarization and citizen engagement during the US and French presidential campaigns. We use the Twitter Gardenhose collection to filter tweets based on keywords around a 50-day window, from March 19, 2012 to May 8, 2012 for the French election and September 19, 2012 to November 8, 2012 for the US Election, particularly focusing on on engagement during the US and French presidential debates on October 3, 2012 and May 2, 2012, respectively. From these data, we constructed partisan alignments based on hashtag usage and retweet networks. We found evidence of more stark political polarization in the French case, while the US case demonstrated less partisan division. This study elaborates commonalities and contrasts in the use of a major social medium by citizens in contexts that differ in political culture and language but feature similar ideological divides, electoral politics, and campaign contexts. We conclude by discussing the implications of computational social science and "big data" in communications, comparative politics, and political sociology.