In an unexpected confluence of events, the January 6 insurrection at the U.S. Capitol unfolded while Gallup was in the midst of conducting a public opinion survey. This serendipitous timing offered researchers a unique opportunity to examine the impact of this significant political event on American public opinion, particularly regarding support for the Republican Party and Donald Trump.
An analysis of this data, published in the journal Electoral Studies, revealed that despite the gravity of the situation, the Republican Party retained a substantial portion of its support base. The findings suggest that Republican voters do not tend to significantly retract support from politicians who violate democratic norms.
“While before the 1990s democracies typically died through military coups, since 2000 four out of every five instances of democratic decline can be attributed to democratically elected leaders undermining the very institutions established to ensure their accountability (a process commonly called ‘executive aggrandizement,’ or more loosely ‘democratic backsliding’),” explained study author Sam van Noort, a lecturer in Public and International Affairs at Princeton University.
“What is interesting about these processes is that they tend to be quite slow, so that voters typically still have at least one, and often several, relatively free and fair elections to vote an autocrat-to-be out of office, before democracy truly dies. Contemporary democratic backsliding can thus often arguably be halted if voters are willing to vote an undemocratic politician out of office.”
“The big question is if voters generally do punish undemocratic behavior by politicians in sufficient numbers to cause his/her electoral defeat. Recent survey experimental evidence in the United States has suggested that American voters do not typically vote undemocratic politicians out of office,” van Noort said.
“The problem with these studies is that they typically rely on relatively modest treatments (e.g., a politician that takes some polling stations away in areas that tend to vote for the opposition) and are hypothetical (which means that it is unclear whether voters would not react differently in the real world when democracy is truly at stake).”
“My study tries to alleviate these weaknesses of existing studies by studying the effect of a novel quasi-experiment created by the fact that Donald Trump’s incitement of the January 6 attack on the U.S. Capitol coincided with a public opinion survey that samples respondents using random digit dialing.
“Because any individual with a U.S. phone number had an equal probability to be sampled just before and just after January 6, 2021, the people who were surveyed just before (say, on January 5, 2021) and just after (say, on January 7, 2021) the event occurred are arguably comparable to each other on all factors besides the undemocratic acts of Donald Trump and his supporters on January 6, 2021,” van Noort explained.
The survey was part of Gallup’s regular “Mood of the Nation” series, and the sample for the current analysis included 1,023 American respondents. The key dependent variables were respondents’ self-reported political affiliation (whether they identified as Republican, lean Republican, Independent, lean Democratic, or Democrat) and their approval or disapproval of Donald Trump’s job performance as president.
Van Noort found only a modest decrease in public support for the Republican Party and Trump in its aftermath. In the 1.5 weeks following the insurrection, the Republican Party’s support decreased by approximately 11 percentage points, indicating that about 78% of those who had identified as Republicans prior to January 6 remained loyal to the party. This reduction in support, while notable, was not as significant as one might expect given the circumstances.
Further analysis revealed that the initial modest loss in support was transient. By February 2021, the Republican Party had recovered to 93% of its pre-insurrection support level.
Similar to the findings on party identification, Trump’s favorability experienced a modest drop from 42.5% to 37.9% in the weeks following the insurrection. Yet, this decline was short-lived, as Trump’s favorability ratings eventually rebounded, indicating that even this direct challenge to democratic norms by a sitting president did not result in a lasting penalty from his base.
The findings suggest that a significant fraction of the electorate may prioritize partisanship over democratic values when the two come into conflict. “For the far majority of Republican Party supporters, even a Republican president inciting an insurrection to overturn the results of a free and fair election — arguably a most-likely case — is insufficient to say in an anonymous phone survey that they no longer support the party/Trump (let alone take costly action to protect democracy),” Van Noort told PsyPost.
“There are many questions that future research can and should address,” the researcher explained. “Arguably the most important ones are: (1) whether the electorate would react differently to different types of democratic backsliding treatments; (2) what background characteristics determine whether someone does or does not change his/her political support away from a co-partisan politician in reaction to undemocratic behavior by that same politician; and (3) how supporters of the Democratic Party would react to similar anti-democratic behavior by co-partisans.”
The study, “How does American public opinion react to overt anti-democratic behavior by politicians? Quasi-experimental evidence from the January 6 insurrection,” was published in the December 2023 issue of Electoral Studies.