This paper sets out to interpolate and predict state-level polling at the daily level by employing a dataset of over 500GB of political tweets from the final months of the 2012 presidential campaign.
This article concerns whether social media is a valid indicator of political behavior. As the authors discuss, there is considerable debate about the validity of data extracted from social media for studying offline behavior. To address this issue, they show that there is a statistically significant association between tweets that mention a candidate for the U.S. House of Representatives and his or her subsequent electoral performance. Accordingly, they demonstrate this result by utilizing an analysis of 542,969 tweets mentioning candidates selected from a random sample of over three billion, as well as Federal Election Commission data from 795 competitive races in the 2010 and 2012 U.S. congressional elections. This finding persists even when controlling for incumbency, district partisanship, media coverage of the race, time, and demographic variables such as the district's racial and gender composition. Their findings show that reliable data about political behavior can be extracted from social media, and forward that more conditional atmospheres may be just as rewarding for research.
This study could be adapted by researchers looking to understand the relationship between online behavior and public perceptions of anti-terrorism and other related policies. The association of tweets regarding government policies could be proxy for understanding public support for those policies.