One of the dominant narratives explaining the election of Donald Trump in 2016 states that he was elected on the backs of white, working-class Americans who felt “left behind” by Washington. Through statistical methods and validated survey data, Jon Green ’14, a doctoral student in political science at The Ohio State University, is demonstrating otherwise.
In a forthcoming study titled “The Differential Effects of Economic Conditions and Racial Attitudes in the Election of Donald Trump,” Green, alongside New York-based researcher and writer Sean McElwee, reached two major conclusions.
First, survey responses regarding race very strongly explain white voters’ preference for Hillary Clinton or Donald Trump in the 2016 presidential election.
Second, economic distress — the variable that political commentators have argued led white, working-class voters to vote for Trump — was strongly associated with the lack of voting among people of color.
“There are a lot of takes [by pundits] out there trying to show evidence that economic distress is what brought us Donald Trump, and I think a lot of folks who have really invested in that argument have been looking for that evidence in the wrong places,” Green said.
Variation in the white vote can be explained by racial attitudes, which is the term Green and McElwee’s study uses for results of survey questions that measure levels of empathy and resentment toward other races as well as opinions on the role race plays in politics. Much of the focus of academic studies about the 2016 election have looked at the two-party vote choice of whites.
“If you want to explain voter participation among other racial subgroups of the electorate, we find evidence that economic distress led people of color to not vote in the first place,” Green said. “So there’s an economic distress story to be told about the 2016 election, but the data does not support the ‘left-behind’ hypothesis and the J.D. Vances of the world who would say that it is economic distress among white people in rural areas that lead them to vote for Donald Trump. We don’t find that to nearly the same degree that we find it elsewhere.”
Green was referencing Ohio-born author J.D. Vance, whose memoir Hillbilly Elegy climbed the bestseller lists following the 2016 election.
One reason why Green feels he was able to get a more accurate picture of the 2016 election data is timing. He used the 2016 Cooperative Congressional Election Survey, a 50,000-person stratified survey administered by YouGov, a market research and data analytics firm. As of Feb. 2018, this survey had been vote-validated — meaning the responses recorded in the survey had been confirmed with voting records from the election — making the data trustworthy.
Green and McElwee combined this data with economic indicators from the IRS at the ZIP code and county levels.
“This is a question we were able to ask because we waited and were publishing later in the game … why we’re able to add value at this point in time to that discussion is just the data that was necessary to do our analysis just wasn’t available,” Green said.
As for interpreting this data, Green says it raises some interesting questions for the 2018 midterms.
“One thing our paper does speak to … is a real polarization in the United States on racial attitudes,” he said, “and I think it’ll be really interesting to see the extent to which that carries over into a non-presidential context … we don’t know yet the extent that’s going to carry over when Donald Trump isn’t on the ballot.”
Green said it’s harder to see how economic distress might play into these upcoming midterms.
“Normally, you would expect economic distress to punish the party that’s in power,” Green explained, “and in 2016 that was the Democrats … so we don’t know if the trend is going to flip if the same economically distressed voters are going to stay home. I just don’t have an answer to that.”
Beyond the midterms, Green hopes his paper will encourage researchers to look beyond just the preferences of white voters.
“In the 2016 election context, people got really good at explaining white voter two-party vote choice and I think it’s the methodological move we take to have a multi-category outcome,” he said.
This means including more objective economic indicators like Green does in his study, to better account for factors like economic distress.
Once it is converted to the journal’s style and format, Green’s forthcoming paper will be published online in Perspectives on Politics, part of the American Political Science Association. The article will come out in print sometime next year.