Tweets and votes: A study of the 2011 Singapore General Election

This study focuses on the uses of Twitter during the elections, examining whether the messages posted online are reflective of the climate of public opinion. Using Twitter data obtained during the official campaign period of the 2011 Singapore General Election, we test the predictive power of tweets in forecasting the election results. In line with some previous studies, we find that during the elections the Twittersphere represents a rich source of data for gauging public opinion and that the frequency of tweets mentioning names of political parties, political candidates and contested constituencies could be used to make predictions about the share of votes at the national level, although the accuracy of the predictions was significantly lower that in the studies done in Germany and the UK. At the level of constituency the predictive power of tweets was much weaker, although still better than chance. The findings suggest that the context in which the elections take place also matters, and that issues like media freedoms, competitiveness of the elections and specifics of the electoral system may lead to certain over-and under-estimations of voting sentiment. The implications for future research are discussed. © 2012 IEEE.