Data from Theory of Ideology Predicts Support for Trump & Antidemocratic Action

• In Event: (Anti)Democracy

Abstract

James Madison famously warned of the threat posed to democracy by the “violence of faction” (Federalist 10 1787). In recent years, Madison’s fears have become manifest in the United States and several other western countries (e.g., Brunello 2018). Actions such as the January 6th, 2021 attack on the US capitol, for example, appear to be motivated in part by the extreme devotion of some Trump supporters who view what others consider assaults on democracy as permissible or justified. Defending democracy in the United States and globally may involve determining how to restrain destructive ideological narratives which convince people that particular actors are so righteous, and their enemies so evil, that victory must be achieved at all costs. This crucial challenge comes at a time when the scientific understanding of political ideology is advancing. To illuminate a path forward, some political psychologists have proposed a focus on elaborating the general structure of ideological belief systems to understand how they can lead people to reject democratic norms and act violently (e.g., Zmigrod 2022).

In response to these calls, we refine and test a theory of the psychological structure of ideological beliefs (e.g., Roseman, Rudolph, Redlawsk, Mattes and Burgess 2021; Roseman, Rudolph, Steele and Katz 2021). This ideological narrative theory (INT) proposes that ideologies are often structured as a canonical narrative that includes five belief components. Much like the narratives of good vs. evil commonly depicted in literature and film, political ideologies tend to invoke a story about an ingroup and outgroup that are in conflict (an identificational component) taking opposing actions (a behavioral component) that would lead to drastically different outcomes (an evaluative component), one of which is highly desirable while the other is ruinous. Ideological narratives also involve claims about the relative morality of the actors, actions, and outcomes (a normative component) and various arguments, anecdotes, and “evidence” demonstrating that the narrative is true (an explanatory component). These beliefs may be strengthened by belief-relevant emotions (e.g., hope for the outcomes; ingroup pride; anger and contempt toward opponents). The five belief components and relevant emotions form a dramatic story that can garner strong commitment and facilitate extreme action.

We tested this theory by looking at support for antidemocratic action and for Donald Trump in the 2020 US Presidential Election. Surveys were fielded in three samples of registered voters: Sample 1 (n = 1200, nationally representative) in the month leading up to the election; Sample 2 (n = 1000, nationally representative) with waves immediately before and after the election) and (Sample 3, n = 287) in the hours and days immediately following the January 6th, 2021 attack. The surveys measured the extent to which respondents endorsed each of five belief components relevant to a pro-Trump narrative and (1) thermometer favorability of Trump and Biden as well as vote choice; (2) extreme antidemocratic behavior including support for a) Trump refusing to leave office if he lost the election; b) blocking some citizens from voting to elect a “better” President; and c) using violence to advance your party’s political goals.

Results from the two nationally representative samples revealed that greater endorsement of each belief component independently predicted higher feeling thermometer ratings for Trump, even after controlling for party ID. Endorsement of several of these belief components also predicted prospective voting for Trump (Sample 1) and actual voting for Trump (Sample 2) after controlling for party ID, with different subsets of components emerging as significant predictors in each sample. Because the belief components are theorized to cooccur in the form of a narrative, we also tested whether simultaneously endorsing multiple components increased support for Trump. Results from both samples showed that as the number of endorsed components increased, so did thermometer ratings for Trump and voting for Trump in the election, with highest thermometer ratings and pro-Trump votes occurring among respondents who endorsed all five components.

In addition, greater endorsement of the pro-Trump belief components both independently and simultaneously predicted greater support for Trump refusing to leave office, even after controlling for party ID. Importantly, this pattern was revealed before the 2020 election (Sample 1) and also immediately after the January 6th attack (Sample 3) when Trump had already engaged in multiple attempts to remain in office after his electoral loss. In sum, these results show support for the INT in multiple contexts and across multiple samples. Further understanding the psychological structure of ideological beliefs can help identify and perhaps affect support for extremist candidates and parties as well as antidemocratic actions.

Authors

• Ben Rudolph, Rutgers University

• Ira Roseman, Rutgers University, CCAS