Electoral Threat and the Impact of Interparty Contact on Affective Polarization

• In Event: (De)Polarization?

Abstract

Does electoral threat condition the effectiveness of interparty contact on affective polarization? Increasing evidence suggests that interparty conversations, both political and non-political, can reduce outparty animosity. However, much of this research has been conducted under conditions in which politics was less salient and when partisan identities might not be as intense, both of which are also conditions that increase the likelihood that this form of contact occurs in reality. In this paper, we seek to understand the extent to which feelings of electoral threat, which are directly tied to expressive partisanship and willingness to vilify the other side, condition the effects of interparty contact on affective polarization. We theorize that electoral "losers," those whose party lost the presidential election, will be more likely to vilify the other side, leading them to be less likely to improve their outparty attitudes after interparty contact. In contrast, "winners," those who identify with the party that won the presidency, will be less likely to vilify the other side and more likely to improve their attitudes toward the outparty after engaging in interparty contact.

We test these expectations with a pre-registered interparty contact experiment fielded immediately following Joe Biden's inauguration as President of the United States. We use a novel block-randomized experimental design in which individuals within interparty partnerships were randomly assigned to have a conversation about the election (contact) or to write a short essay about the election (no contact). Conversations occurred on an innovative online chat software. We measure within-subject change in out-party animosity, social polarization, perceptions of election integrity, and democratic values. Importantly, we examine heterogeneous treatment effects based on feelings of electoral threat (Republicans) or reassurance (Democrats).

We test the scope of our theoretical contribution with a host of robustness checks. Of note, we examine (1) heterogeneous treatment effects based on pre-treatment preference for interparty contact, (2) the durability of treatment effects 3 days after contact, and (3) different measures of outparty animosity, addressing concerns recently raised in the literature, suggesting that affective polarization has been overestimated due to instrumentation error.

All told, our results speak to the growing conversation about the limits (and possibilities) of interparty contact as a means of reducing tensions between Republicans and Democrats in the United States.

Authors

• Erin Rossiter

• Taylor Nicole Carlson, Washington University in St. Louis