Social Media, News and Political Information during the US Election: Was Polarizing Content Concentrated in Swing States?

US voters shared large volumes of polarizing political news and information in the form of links to content from Russian, WikiLeaks and junk news sources. Was this low quality political information distributed evenly around the country, or concentrated in swing states and particular parts of the country? In this data memo we apply a tested dictionary of sources about political news and information being shared over Twitter over a ten day period around the 2016 Presidential Election. Using self-reported location information, we place a third of users by state and create a simple index for the distribution of polarizing content around the country. We find that (1) nationally, Twitter users got more misinformation, polarizing and conspiratorial content than professionally produced news. (2) Users in some states, however, shared more polarizing political news and information than users in other states. (3) Average levels of misinformation were higher in swing states than in uncontested states, even when weighted for the relative size of the user population in each state. We conclude with some observations about the impact of strategically disseminated polarizing information on public life. COMPUTATIONAL PROPAGANDA AND THE 2016 US ELECTION Social media plays an important role in the circulation of ideas about public policy and politics. Political actors and governments worldwide are deploying both people and algorithms to shape public life. Bots are pieces of software intended to perform simple, repetitive, and robotic tasks. They can perform legitimate tasks on social media like delivering news and information—real news as well as junk—or undertake malicious activities like spamming, harassment and hate speech. Whatever their uses, bots on social media platforms are able to rapidly deploy messages, replicate themselves, and pass as human users. They are also a pernicious means of spreading junk news over social networks of family and friends. Computational propaganda flourished during the 2016 US Presidential Election. There were numerous examples of misinformation distributed online with the intention of misleading voters or simply earning a profit. Multiple media reports have investigated how “fake news” may have propelled Donald J. Trump to victory. What kinds of political news and information were social media users in the United States sharing in advance of voting day? How much of it was extremist, sensationalist, conspiratorial, masked commentary, fake, or some other form of junk news? Was this misleading information concentrated in the battleground states where the margins of victory for candidates had big consequences for electoral outcomes? SOCIAL MEDIA AND JUNK NEWS Junk news, widely distributed over social media platforms, can in many cases be considered to be a form of computational propaganda. Social media platforms have served significant volumes of fake, sensational, and other forms of junk news at sensitive moments in public life, though most platforms reveal little about how much of this content there is or what its impact on users may be. The World Economic Forum recently identified the rapid spread of misinformation online as among the top 10 perils to society. Prior research has found that social media favors sensationalist content, regardless of whether the content has been fact checked or is from a reliable source. When junk news is backed by automation, either through dissemination algorithms that the platform operators cannot fully explain or through political bots that promote content in a preprogrammed way, political actors have a powerful set of tools for computational propaganda. Both state and non-state political actors deliberately manipulate and amplify non-factual information online. Fake news websites deliberately publish misleading, deceptive or incorrect information purporting to be real news for political, economic or cultural gain. These sites often rely on social media to attract web traffic and drive engagement. Both fake news websites and political bots are crucial tools in digital propaganda attacks—they aim to influence conversations, demobilize opposition and generate false support. SAMPLING AND METHOD Our analysis is based on a dataset of 22,117,221 tweets collected between November 1-11, 2016, that contained hashtags related to politics and the election in the US. Our previous analyses have been based on samples of political conversation, over Twitter that used hashtags that were relevant to the US election as a whole. In this analysis, we selected users who provided some evidence of physical location across the United States in their profiles. Within our initial sample, approximately 7,083,691 tweets, 32 percent of the total