Does the early bird move the polls?

Social media tools such as Twitter and Facebook have been hailed as politically transformative communications technologies, on the level of previous politically significant technological innovations, such as radio and television. In this paper we estimate the impact of the use of the social media tool 'Twitter' by American politicians on changes in public opinion of those politicians over time, and we find little evidence that the political use of Twitter has eitther a positive or negative impact on public opinion. We run simple regressions of changes in widely used U. S. public opinion polls, estimating the association between favorable impressions of selected politicians and the social media usage levels (number of 'Tweets') of those politicians. In order to get a reliable estimate, we ran the same regressions for public opinion polls conducted by three different companies (CNN/Opinion Research, Gallup, and ABC/Washington Post) and we find that the estimated coefficients are not significant for any of the tested opinion polls.