The Role of Social Media Platforms in Forecasting Elections: A Comparison of Twitter and Facebook

Abstract

Prior literature shows that social media could be used to forecast political elections. Most studies have focused on a single social media platform, and few studies have explored the use of social media data across multiple platforms to make election predictions. Though candidates’ personal attributes have also been suggested as critical factors affecting election results, there has been little research into the interacting effect of social media and candidate attributes in predicting elections. To address the research gap, this paper investigates the role of two different social media platforms, Twitter and Facebook, in forecasting the 2019 Finnish parliamentary elections, and how candidates’ political experience moderates the role of the two platforms in predicting elections. The findings show that both the number of Facebook likes and the number of Twitter followers are associated with the election outcome positively. Political experience of candidates moderates the association between the number of Facebook likes and election outcomes as well as the association between the number of Twitter followers and election outcomes. This research adds to the discussion on how social media can predict election results. It considers both different social media platforms and the roles of the candidate attributes.