The American Upvoter: The Internet and Political Participation

The internet has newly amplified and distorted the many voices and pressures present on the stage of American life, and has thus fundamentally changed the political landscape of the United States. Political action, thanks to the connective power of online communication, is easier to organize than ever before, benefiting both benign and disruptive actors. However, our understanding of these changes often relies on incompatible or outdated theoretical perspectives and methodological approaches. This project aims to remedy this problem by offering three studies which approach this topic in an appropriate way.

The first study, “Squaring the Virtuous Circle,” provides one of the few causal estimates of the provision of internet access on individual voter turnout and propensity to engage in the political process. The second and third studies, “Live, Laugh, Lurk” and “I’ll Listen to You Schizo But It’s Time to Take Your Meds,” investigate political discussion on Reddit, and 4Chan, respectively. Together, these two chapters demonstrate an alternative approach to commonplace survey measures of online political engagement by directly measuring the comments of individuals. It does so by matching respondents from the 2008-2009 American National Election Panel to respondents from the main 2008 study. Overall, I find that internet provision to those who lack access causes a 13% increase in one’s likelihood of voting, as well as increased participation in the political process.

The second study, “Live, Laugh, Lurk” finds that, on Reddit, political knowledge is unrelated to the rate at which individuals contribute political content, and that political interest is primarily involved with an initial political post rather than the motivation of further political posts. Additionally, I find that free time and civic skills, core components of the civic voluntarism model of political participation, are unrelated to posting political content.

The third study, “I’ll Listen to You Schizo, But It’s Time to Take Your Meds,” finds that, contrary to my expectations, the novelty and opinion content of a thread’s original post are negatively related to the number of replies it receives. I argue that the hierarchical structure of the data must be modeled in order to properly discern the relationship between a post’s content and the attention it receives.