Despite getting the winner wrong, most of the national polling in the final weeks of the election were fairly accurate, with Hillary Clinton winning between two to three percent of the popular vote. Using this data, a new study found that the clearest predictor for Trump support were racist and sexist attitudes.
Political scientists Brian Schaffner, Matthew MacWilliams, and Tatishe Nteta published a paper showing from survey data that voters most likely to be supportive of racism or sexism supported Trump at high margins,
The two above charts show this correlation. On the left, we see the percentage of white voters with an without a college who voted for Trump. The chart onthe right shows the probability rating of voting for Trump based on economic dissatisfaction, racism, and sexism. In this latter chart, the high end of all these things predict support for Trump, but the racism and sexism lines are more pronounced and nearly identical.
These findings are not exactly new, as noted by Vox:
Several polls found that Trump supporters were more likely to profess negative views of black people, Muslims, and Latinos, as well as concerns that immigrants threaten US values. One telling study, conducted by researchers at UC Santa Barbara and Stanford University shortly before the election, found that if people who strongly identified as white were told that nonwhite groups will outnumber white people in 2042, they became more likely to support Trump.
Another set of studies, conducted by researchers Carly Wayne, Nicholas Valentino, and Marzia Oceno, found that measures of benevolent sexism — meaning more traditional, chivalrous views of women and men’s proper roles in society — didn’t correlate closely with support for Trump. But measures of hostile sexism did, suggesting that sexism in support of Trump seems to be more about hostility toward women than old-fashioned views on gender roles.
Still, correlation does not equal causation. Possibly the most important finding in this study, however, is how it affirms other data showing that lack of education is also a key predictor of Trump support. In fact, the first section of the paper deals with the “education” gap amongst white voters.
It does not seem like a stretch of credibility to claim that “racism” and “sexism” are the products of ignorance as much as they are the product of hate. It also seems like a fair assumption that those with little education are much easier to dupe with lies and false stories than those who’ve graduated from college.
While there is much worth debating about the value of post-secondary education, the current system does place an emphasis on critical thinking skills and the value of good sourcing. Of course, college-education folks are just as susceptible to confirmation bias as the non-college educated, so this could be an area where further study is required.
Still, this seems to underscore the importance of both education and sources–meaning the news media–that are trustworthy and accurate. Without these things, leaders are able to lie more easily, scapegoat people or demographic groups, and, in general, makes America an angrier and more hostile place.
What do you think? Are these findings on racism and sexism a problem of education? How do you think it can best be addressed? Share your thoughts and ideas in the comments below.