Regional differences in perceptions of the threats to U.S. democracy

A Cornell-IOPGA poll examined Americans’ attitudes toward various alleged threats to the republic with special emphasis on two swing districts; here’s how the results broke down via the American Nations model

By Douglas L. Kriner and Colin Woodard

Nearly a year ago, Cornell University’s Institute of Politics and Global Affairs organized a detailed survey examining the political views of American voters generally and of those in two battleground congressional districts with special focus on attitudes toward various alleged threats to U.S. democracy and competing narratives about how to counter them. The survey found residents of the swing districts – in central Michigan and south Texas – were more concerned about threats presented by President Trump and his supporters than the country as a whole.

We were both at Cornell’s Manhattan launch event – the 2022 State of Democracy Summit – Kriner as co-author of the survey, Woodard to present regional and cultural interpretations of what the survey results showed. We both found the results important and, in some cases, unexpected – more on that in a moment – but we were also curious to what extent regional cultural influences might be at play, both in explaining the two swing district’s results and in segmenting the national sample.

Today we’re able to present some answers.

Kriner, Clinton Rossiter Professor in American Institutions at Cornell’s Department of Government and faculty director at Cornell-IOPGA broke down the national survey responses by region using Woodard’s American Nations model (which is explained here if you are not familiar with it.) We then compared the results between the regional cultures, focusing on questions about various threats to U.S. democracy, and also with the swing districts. (For details on the methodology, the survey, confidence bars, end notes and other details see our “Under the Hood” appendix page.)

First, there are indeed substantial regional differences in how Americans perceive the source and seriousness of threats to the republic, with the Deep South and Left Coast often appearing as outliers on opposite sides of the divide.

Take January 6th, for instance. Respondents were asked if they thought the attack on the U.S. Capitol – which has been defined as a coup attempt by the Cline Center for Advanced Social Research – represented a minor or major threat to democracy or was not a threat at all. Nationwide a narrow majority – 51% — said it was a major threat. But there was 22-point gap between respondents in the Deep South (42% of whom saw the attack as a major threat) and the Left Coast (64%). Greater Appalachia – the region most supportive of Trump in both of his presidential bids – also discounted the danger, with only 46% regarding it as a major threat, while residents of Tidewater, the Midlands, and New Netherland were more concerned than the national average. (1)

Was the January 6th attack on the U.S. Capitol by a mob of pro-Trump voters a major threat? Credit: Douglas Kriner

The regional gap was also great among Jan. 6th threat deniers, with 20.2 percent of respondents in Greater Appalachia saying the attacks represented no threat at all compared with just 5.6 percent of Left Coasters and 9 percent of Yankeedom residents.

Respondents were also asked to assess the threat posed by 19 state legislatures trying to change voting laws to prevent people of color from voting or making it harder for them to vote and, again, 51% of Americans overall said this was a major threat. In the Deep South, Greater Appalachia, and El Norte only 46 to 48 percent shared this view while in the Left Coast, Tidewater, and the classic swing region of the Midlands the rate topped 60 percent. Similarly, the percentage of Greater Appalachians who regard this as no threat at all (17.9) is more than double that of Left Coaters and Yankees (both 7.6.) (2)

19 state legislatures have tried to change voting laws to make it harder to vote or to prevent people of color from voting. Is it a threat to democracy? Credit: Doug Kriner.

Views were reversed when respondents were asked about the threat of voter fraud – which is not supported by empirical evidence – and the alleged effort by “social media platforms like Facebook and Twitter” to “intentionally censor political viewpoints that they disagree with.” Social media censorship was broadly dismissed as a serious threat by respondents from Left Coast (28%) New Netherland (34%) and El Norte (35%), but was the majority position in the Deep South (52%). (3) Voter fraud was rejected as a serious concern by Americans as a whole (only 37% thought otherwise) and almost every region. The exception: Deep South, where half of respondents thought it a major threat, followed by Greater Appalachia with 41%, which was still more than double the level in the Left Coast (19%.) (4)

States are making it easier to vote via absentee and mail voting. Is this a threat to democracy? Credits: Doug Kriner.
Social media platforms are intentionally censoring political viewpoints they disagree with. Is this a threat to democracy?

Regional polarization followed different lines with a question about the threat posed by  “attempts to erase American history” via protests, vandalism and calls to remove “statutes and memorials to our past.” The question was a reference to the efforts to remove monuments to Confederate heroes, Christopher Columbus, and other statuary dedicated to figures in disrepute. Here the primary gulf was between East and West, perhaps because fights over the legacy of the Civil War have less valence in parts of the country that were on the periphery of that conflict. Fifty percent of Americans said statue removal efforts were a major threat, a position held by a majority of respondents in Greater Appalachia (56%), Deep South (55%), the Midlands (54%) and Yankeedom (51%). This was a minority position in Left Coast (32%), Far West (43%), and El Norte (45%). But it was also rejected by respondents in two eastern seaboard regions: New Netherland (41%) and Tidewater (34%), a region that’s been rapidly devolving in recent decades due to a combination of its small size and the large presence of the federal government in the District of Columbia and the Hampton Roads region, the latter home to the world’s largest naval base. (5)


There have been attempts to erase American history through vandalism or protests to remove statues and memorials to our past. Is this a threat to democracy? Credit: Douglas Kriner.

That said, we found the two swing districts really do stand out in many ways, not just from the rest of the nation but from the regional cultures they respectively belong to. Though they stand on opposite ends of the country and in two very different cultural environments their residents showed a much acuter concern with anti-democratic threats.

The districts were Michigan’s Eighth, a section in the central part of the state currently represented by Democrat Dan Kildee, and Texas’s Fifteenth, a spaghetti lot of a district whose residents primarily live either around border city of McAllen in the Lower Rio Grande Valley or more than 200 miles north in the eastern suburbs of San Antonio and where the incumbent is Republican Monica de la Cruz. They were chosen because they are competitive districts that have switched back and forth between parties in recent presidential elections and where voters still sometimes ticket split.

The survey asked about ten threats to democracy. Respondents in both districts were more concerned than Americans at large about threats posed by the Jan. 6 attacks, harassment of elected officials, efforts by state legislatures to prevent people of color from voting and members of Congress trying to overthrow the 2020 election.  TX-15 was particularly concerned, with 66 percent considering Jan. 6 a major threat, fifteen points higher than the national average, and 56% worried about the Freedom Caucus members trying to overturn the election, a level 14 points higher than the nation. (Interestingly, both districts are also more supportive of implementing voter ID laws to protect elections than Americans as a whole (thought the difference in MI-8 was smaller and not statistically significant), and in TX-15 the 10-point difference was driven mostly by higher levels of support by Democrats.) (6)

Turns out these two districts also stood out in these issues in comparison with their respective regional cultures too, though by smaller margins. TX-15 respondents saw all ten potential threats to democracy as more concerning than the overall El Norte sample, by 20 to 30 points in many instances.  They were also significantly more likely to support voter ID laws than other residents of El Norte.  MI-8 respondents ranked 9 out of 10 threats as more threatening than their Yankee counterparts, but most only by a 5 to 10 percent margin. The differences remained similar even in statistical models that controlled for partisanship, education, age and race. (7)

This suggests there is something “special” about the swing districts in regards to democracy protection, at least in Yankeedom and El Norte. The patterns seen in MI-8 may well have parallels in ME-02, WI-03, NY-19 and 24, and NH-1. And what’s true in TX-15 may be even more true of TX 23, 28, and 34, the Lower Rio Grande Valley districts. These are places where Donald Trump made some of his greatest gains in the whole country between 2016 and 2020, propelled by a massive region wide defection of previously reliable rural Democratic voters to Trump. Given the key role these voters would play in any scenario resulting in Texas “turning blue” – and effectively locking Republicans out of the Electoral College — everyone should be paying close attention to what they think, especially the Tejanos of South Texas.

Douglas L. Kriner is Clinton Rossiter Professor in American Institutions in the Department of Government at Cornell University and faculty director of Cornell’s Institute of Politics and Government Affairs. Colin Woodard is director of the Pell Center’s Nationhood Lab.