Commentary – New Book on Rural America Started with a False Conclusion, Then Looked for Evidence
by Nicholas F. Jacobs, The Daily Yonder
March 6, 2024
A new book on “white rural rage” argues that rural Americans are the most racist, xenophobic, conspiracist, anti-democratic, and violent “geodemographic” subgroup of Americans out there. The authors, Tom Schaller and Paul Waldman, say that their book is driven by data and that even if you don’t like their conclusions the evidence is clear. The problem with this book is not a lack of citations; they “bring the receipts,” as they like to boast. The problem is that those receipts belong in the trash.
Scholars have repeatedly shown that there is a difference between rural and urban voters, and it’s not rage, racism, or homophobia. But before I explain more about that rural difference, let’s look at the many ways Schaller and Waldman misuse data to support their claim that rural Americans are the nation’s biggest threat to democracy.
A Salacious Title in Search of Facts
I’ve reviewed every publicly available survey and poll they use to stereotype, marginalize, and ultimately demean the lives and beliefs of rural Americans. The book reeks of tell-tale signs of being written first and finding facts second. Only after they settled on a salacious title, it seems, did they go out and try to find what they already agreed to see, with little to no attention paid to whether any of it was true. Although they are sure to sell books, their selective use of sketchy data will worsen our efforts to rebuild rural communities and close a rural-urban divide that threatens our democracy, regardless of where you live.
This is not normal academic quibbling about numbers and definitions. The faults I describe below lead the authors to drastically different conclusions that more rigorously conducted research demonstrates. Where rural politics scholars are increasingly attentive to the idea that “rage” is not the same as economic anxiety, community pride, and a sense of place, Schaller and Waldman simply want us to write off rural America as the land of radical extremism. By the time they even get to some warm-hearted conclusion on empowering rural community, they’ve so misrepresented and maligned rural people that nobody in the countryside can (or should) trust them, and nobody inside urban America will find it worthwhile to care about some wasteland of hate.
Below I detail the four most significant problems with the underlying data presented throughout the book. I offer a more comprehensive list of each survey and poll they use here, and briefly describe my concerns below.
A Litany of Problems
The first problem is that the vast majority of data lacks any consistent definition of what they mean by the word “rural.” This hole gives the authors license to pick and choose various points from whatever surveys they want with no regard for what those studies say about rural people – if, in fact, the studies care to define rural in the first place.
Consider, as just one of several examples, a method by the polling firm IPSOS, which informs work from the Public Religion Research Institute (PRRI) that Schaller and Waldman use to prove that rural folks are distinctively drawn to QAnon conspiracies. IPSOS defines rural as any resident living in a county that is not a part of any Metropolitan Statistical Area (a place with a dense urban center of 50,000 people or more). The Census, however, estimates that 54% of all rural people actually live inside those “metropolitan” areas — areas excluded by the IPSOS definition. Consequently, this result about “rural America” is drawn from a survey that excludes a majority of rural residents. One wonders if the authors cared to think about what this actually means for their results, or if the headline grabbing “finding” is all that mattered.
Second, even if we were to have a consistent definition, Schaller and Waldman seem to willfully neglect any concern over sample sizes. More often than not, the surveys that inform this “data-driven” account simply do not get enough respondents who are even from rural areas, however defined. Yes, survey outlets will report out what “rural” individuals said, but a closer look shows that those rural estimates often draw on just a few hundred people. Small sample size introduces increasing amounts of uncertainty and error. The smaller the sample, the larger the margin of error. When these margins get too big, it’s impossible to use the answers with any statistical certainty.
Consider a poll by the Institute of Politics, which Waldman and Schaller reference to show that rural Americans are the most likely to “take up arms against the government.” That claim depends on just 220 rural residents and 290 city residents. As such, if 35% of rural residents agree, and 29% of city residents agree, probabilistically they are indistinguishable from one another.
Related Story
Or consider another poll by Marist University, which Schaller and Waldman say proves the point that rural residents are more likely to believe in nonexistent voter fraud. That conclusion depends on a grand total of 167 rural individuals. Not only are the margins of error too big to be meaningful, you might wonder, can you even get a representative portrayal of rural America with that many people.
No. And that’s the third problem. Because most surveys are done to get a representative picture of the national population, even when they do have adequate sample sizes, seldom do the rural respondents in the poll actually represent the demographics of rural America.
Two of the most egregious cases in the book are when the authors rely on a poll of just two states to make claims about birtherism in rural America and when they present a survey of residents in just nine states, which collectively represent just 31% of the rural American population.
Samples That Don’t Reflect Rural Demographics
In the last seven years, I’ve surveyed over 25,000 rural residents. My experience is that the first to respond and fill that “rural” quota are older and more conservative than average. Getting young rural folks is hard, and most surveys might only have a handful. Alas, the vast majority of studies Schaller and Waldman cite do not report out demographic statistics on their rural populations. Did they even care if they represented “rural” communities?
This demographic imbalance exists even in high-quality surveys, because most surveys are designed to make claims about the national population. Surveys weights are therefore used to adjust samples to mirror national demographic patterns. In the case of the American National Election Study, for example, this means that any estimate about rural residents inflates the voices of the most elderly residents (65+), because ANES weights create an incorrect demographic snapshot of rural America, with 30% more elderly residents than what the U.S. Census says should exist.
Only two surveys in the entire book conform to basic standards of survey research and even attempt to try and present an accurate picture of rural America: a 2017 study from The Washington Post and Kaiser Family Foundation (1,070 rural residents) and a 2018 report from the Pew Research Center (2,085 rural residents) — a study, by the way, that shows that majorities of rural Americans believe that “white people benefit from advantages in society that black people do not have,” that “there are still significant obstacles that make it harder for women to get ahead than men,” that there are, indeed, “situations in which abortion should be allowed,” and who reject the idea that a non-white, majority country would be “bad” for America. Oh, the rage!
Of course, not even these are perfect. Because it focuses on rural communities, the WAPO/KFF study only includes 303 urban and 307 suburban respondents – small samples. Consider what this means when interpreting who thinks that “immigrants today strengthen our country,” one of the questions Schaller and Waldman use to show just how xenophobic rural people allegedly are. Once you account for the margin of error (which Schaller and Waldman never report once in the book), rural residents actually believe the same thing as suburban residents; indeed, one question down (remember cherry-picking?), the survey shows that rural residents are just as likely as urban residents to say that “most immigrants coming to the U.S. in the last 10 years are doing enough to adapt to the American way of life.” Can you find the rural rage?
What’s Rural Got to Do with It?
The final problem is the worst one because it exists even if you account for problems of defining rurality and getting large, representative samples. Throughout the book, Schaller and Waldman solely rely on group comparisons – a true indication that their analysis is driven more by a desire to confirm one’s beliefs than to find real evidence for “rural rage.” Nowhere in this book is there any attempt to understand what motivates rural people, in particular, to think one thing or another. Are these really rural problems, or are they the product of some other demographic characteristic such as age, race, or gender? Is the “rage” supposedly felt in rural areas really different than elsewhere? We just cannot tell from group averages alone.
Schaller and Waldman, for instance, are concerned about the rise of QAnon conspiracy theory beliefs. I am too. Using the 2021 American Values Survey from PRRI (the same one that excludes majority of rural respondents), they rest their claims of rural rage on the fact that 24% of rural residents are QAnon believers. The rage is there, they say, because that is disproportionately higher than the 17% of Americans who are rural, as estimated by PRRI (it is possible these are statistically the same, but data to compute the margin of error is not disclosed). Nevertheless, the problem is that, from this simple comparison alone, they conclude that rural is extreme; rural is the hotbed of QAnon; rural is danger land. But the vast majority of rural residents don’t believe in QAnon conspiracies. And there are over three times as many QAnon believers outside rural America. That makes rural America the threat to American democracy?
Consider the same fallacy as it relates to something like vaccine uptake. Black Americans are the least likely to have had a Covid vaccine. Whatever the reason for this phenomenon, no one would jump from that finding to say “the real threat to public health comes from Black Americans’ refusal to take the vaccine. They are the least likely group to get it.” We know that claim is nonsense because unvaccinated Black Americans represent such a small percent of the unvaccinated population (just like rural!). Moreover, low vaccine rates might actually have little to do with race per se, but instead be the consequence of other factors we know are associated with vaccine uptake and race — maybe income or education? If we actually cared about increasing vaccine uptake (or fighting conspiracy theory), we wouldn’t give a second thought to such a shoddy argument.
Instead, the way we would really test why certain groups behave a certain way – be it conspiracies or vaccines — is to compare individuals within and between those groups so that we can account for the multiple differences that they might have in common. Schaller and Waldman themselves never dig this deep, keeping the analysis simplistic and shallow.
But I got some of their data and did the work. Using the individual-level data provided by the 2018 Pew study, for example, I tested Schaller and Waldman’s claims that rural residents are distinctively “xenophobic.” At a group level, 57% (+/- 1.1) of rural residents compared to 35 % (+/- 0.68) of urbanites, agree that a “growing number of newcomers from other countries threatens traditional American customs and values.”
I relied on widely used and understood tools of statistical inference (i.e., a regression model) to understand what best explains a person’s response to that question. Ruralness? Maybe. But maybe they answer that way because of age, sex, level of education, party identification, income, race, some religious beliefs, or all of the above.
Here’s the model’s results. Bottom line: when we include those characteristics in a model that also accounts for where someone lives, geographic differences disappear. Rural residents are statistically no more likely to think “xenophobic” beliefs on account of their rurality. Attitudes here are almost entirely attributable to partisan identity. Being rural adds nothing to the explanation. It is statistically meaningless, because what drives those group averages is solely a function of Republican partisan identity. That is, rural residents are thinking the same way about that issue as urban and suburban residents. You may disagree with that thought, but your disagreement is with Republicans, not rural people. And if the goal of White Rural Rage is to say that the threat to American democracy is emanating from the heartland, the data just do not support that idea.
What Is the “Rural Difference”?
Schaller and Waldman might come back and say that that is a difference without a distinction. Because rural residents have some disproportionate weight at the ballot box, their group level influence is what reverberates throughout American democracy.
But I thought the goal was to understand motivations. Why do rural residents hold these beliefs? Why are they so enraged? What makes them, as a group, more likely to vote for someone like Donald Trump? Motivations explain why people are drawn to some candidates and not others. Motivations are how we build new coalitions and encourage candidates to speak to the issues that are driving illiberalism. Motivations are the thing we have to study if we truly care about the threat to democracy rather than book sales.
Scholars, again, and again, and again, however, have shown that there is something distinct about rural people’s motivations, and it isn’t rage. In my latest book with Dan Shea, we rely on genuinely representative samples of 10,000 rural Americans. And, actually driven by the data and concerned with its quality, we find that there is a distinctive reason why so many rural residents find Trump appealing and have moved towards the Republican Party. They feel unheard. They worry about the future of their community and their kids having to move away. They do so much with so little and they resent it when someone from outside their community steps in and pretends to know what they think (especially when they pretend to have the facts). Far from the stereotypes you find in White Rural Rage, we find that rural folks are overwhelmingly proud of their communities, but feel that elites look down on them, stereotype their cherished ways of living, typecast them in broad strokes – simply put, see nothing, but rural rage.
Nicholas F. Jacobs is an assistant professor of government at Colby College and the co-author, with Daniel M. Shea, of The Rural Voter: The Politics of Place and the Disuniting of America.
This article first appeared on The Daily Yonder and is republished here under a Creative Commons license.
Trego's Mountain Ear 
Leave a comment