Designing for Safe, Fun and Informative Online Cross-Partisan Interactions

Abstract

The past decade in the US has been one of the most politically polarizing in recent memory. Increasingly, ordinary Democrats and Republicans fundamentally dislike and distrust each other, even when they agree on policy issues. Most Americans report believing that the opposing party is a "serious threat to the United States and its people". This extreme partisan hostility has wide-ranging consequences, even affecting how partisans respond to COVID-19 mitigation measures. In this context, this dissertation aims to reduce hostile interactions and attitudes towards ordinary Democrats and Republicans. I argue that we can reduce hostility by leveraging nonpolitical online spaces that cut through the partisan faultlines in uniquely engaging ways. I develop approaches to transform the currently hostile, uninspiring nature of online political interactions into not only a safe experience but also a fun and informative one. I take a mixed-methods approach to studying outpartisan hostility, combining computational social science with design methods. The dissertation progresses from a large-scale exploratory analysis of online political discussions to developing potential designs to reduce online partisan hostility and, finally, to designing and evaluating a fun party game that reduces outparty hostility.

In the first study, through large-scale computational analysis of billions of Reddit comments, I find that nearly half of all political discussions on Reddit take place in nonpolitical communities and that cross-partisan political conversations in these communities are less toxic than those in explicitly political communities. These findings suggest that shared nonpolitical interests can temper online partisan hostility. In the second study, through in-depth qualitative interviews and design probes, I explore approaches to surface these nonpolitical interests and identities during online political interactions on Reddit. I demonstrate that participants are comfortable knowing and revealing shared memberships in nonpolitical communities with outpartisan discussion partners which they expect to be humanizing, potentially reducing the hostility in those interactions. Through the interviews, I find that apart from serious deliberative discussions, participants also engage in light-hearted and casual political interactions where the motivation to simply entertain themselves and have fun. In the final study, drawing on insights from the prior study and extant political science research, I develop an online party game that combines the relaxed, playful nonpartisan norms of casual games with corrective information about Democrats' and Republicans' political views that are often misperceived. Through an experiment, I find that playing the game significantly reduces hostile attitudes toward outparty supporters among Democrats.

Overall, this dissertation demonstrates the potential of using nonpolitical context to facilitate quality online cross-partisan interactions that account for and mitigate the heightened levels of partisan animosity we observe today.