Discussion’s Role in Affectively Polarizing the American Public

• In Event: Partisanship in the Media

Abstract

A central tenant of a democratic society is the duty of citizens to share and to hear conflicting political views. Normatively, discussion contributes to a healthy democracy as a form of participation whereby citizens formulate and express their own views and gather information. In practice, however, it can be difficult and uncomfortable to engage in disagreeable political discussion, especially given the heightened partisan animus that characterizes American's attitudes today. This paradox between the proposed and practiced role of discussion begs the question, what happens when two people discuss politics? Does the exchange achieve its normative goals or have adverse effects?

In this paper, I examine the relationship between discussion and the well-documented hostility that exists for members of the opposite political party. Since partisan identification is a strong social identity, it is a salient characteristic when any two people discuss politics. But to what end? The literature suggests two competing hypotheses. Contact theory suggests that discussion will ameliorate outgroup animus via the learning that occurs through contact with outgroup members, while social identity theory suggests that discussion will only heighten outgroup animus by prompting partisans to engage in motivated reasoning to maintain the comfortable distinction between ingroup and outgroup members.

I test these competing hypotheses with a series of experiments on Amazon Mechanical Turk in which participants engage in real-time, written political discussion using a novel online chat software developed specifically for the academic use of experimentally manipulating the conditions under which individuals engage in discussion. I randomly assign discussion partners to be copartisans or not, ask them to engage in a political discussion, and measure changes in attitudes as a result of the discussion. I expect to find support for one of the competing hypotheses by examining whether the treatment effect of discussing politics with an outgroup member leads to an increase or decrease in outgroup animus.

This paper highlights political discussion itself as an important political phenomenon, and introduces a novel experimental framework for studying the effect of discussion on attitudes and behaviors. Further, this paper addresses a gap in our understanding of the role discussion plays in either harming or healing our already hostile political environment.

Author

• Erin Rossiter