Can we vote with our tweet? On the perennial difficulty of election forecasting with social media

Social media and other “big” data promise new sources of information for tracking and forecasting electoral contests in democratic societies. This paper discusses the use of social media, and Twitter in particular, for forecasting elections in the United States, Germany, and other democracies. All known forecasting methods based on social media have failed when subjected to the demands of true forward-looking electoral forecasting. These failures appear to be due to fundamental properties of social media, rather than to methodological or algorithmic difficulties. In short, social media do not, and probably never will, offer a stable, unbiased, representative picture of the electorate; and convenience samples of social media lack sufficient data to fix these problems post hoc. Hence, while these services may, as others in this volume discuss, offer new ways of reaching prospective voters, the data that they generate will not replace polling as a means of assessing the sentiment or intentions of the electorate.