More tweets, More votes: Social media as a quantitative indicator of political behavior New work by Joseph DiGrazia, Karissa McKelvey, Johan Bollen and Fabio Rojas. We hope this paper can convince your reviewers that social media aren't just spam-infested pools of emoticons. We show that random samples are significant predictors of vote margin in house elections. No sentiment needed -- only raw counts of the names of the candidates. We do it by measuring simply the number of times a candidate was mentioned by full name (first last). Handles and hashtags don't work as well. There will be another paper later about that. The working paper (short!) is here: http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2235423 We'd appreciate your comments -- Karissa McKelvey rissarae.net