Geography and Polarization: Does Political Heterogeneity Polarize Citizens?

• In Event: Interpersonal Networks in Tweets, Conversations, Neighborhoods, and the Pandemic

Abstract

Researchers have found that discussions in politically heterogenous social groups tend to moderate individuals’ political attitudes (Klar, 2014; Klofstad et al., 2013). Such results agree with the so-called “echo chamber” hypothesis in which individuals confronted with information consistent with their attitudes will continue to engage in biased reasoning. It is unclear, however, whether echo chambers can form in larger units of social and geographic organization, such as the neighborhood, municipality, or county. Individuals may have no or few social ties to these broader social groups. What ties they do have may be low stakes, fleeting, and/or occur unpredictably. In fact, preliminary evidence shows that voters are polarized in congressional districts that are ideologically moderate on average (McCarty et al., 2019). In other words, politically heterogenous congressional districts may not moderate individuals’ political attitudes – heterogenous districts might be polarizing them. This suggests a possible “triggering effect.” Citizens in politically heterogenous neighborhoods, municipalities, and counties might feel the need adopt stronger party ID and political attitudes as a defense against consistent interaction with outgroup partisans. These citizens will more often be triggered by interactions with outgroup partisans and thus adopt more extreme political opinions when confronted with attitudes and information that contradict their own political positions – a phenomenon corroborated by experimental research (Nyhan and Reifler, 2010). The presence of a triggering effect in these larger socio-geographic units would be important at a time when scholars, pundits, and citizens are concerned about political segregation and its effects on political polarization. This paper uses ANES data and multilevel modeling to understand the influence of political heterogenous neighborhoods, municipalities, and counties on individuals’ party ID, political attitudes, and political affect.

Author

• Patrick Villalonga, University of Arkansas