Even Among The Wealthy, Education Predicts Trump Support

Last week, my colleague Nate Silver used census data to show that education, not income, determined the shift from Republican to Democratic votes in the 2016 election. It turns out that the exit polls can also help us confirm and expand that thesis.

First, it’s clear from the exit polls that for white voters, every bit of extra education meant less support for Trump.¹ That is, it wasn’t just a matter of attending college or getting a degree. While much has been made of the college and non-college divide (which is stark), Trump actually won whites who earned only a bachelor’s degree by a fairly wide margin. Just as big a gap was between the votes of those who graduated from college and those who went to graduate school. The latter group supported Clinton in much larger numbers.

Second, education matters a lot even when separating out income levels. Trump won by 24 percentage points or more among every single income group of whites without a college education than those with one. In his worst income group among those without a college degree, he did over 20 percentage points better than he did in his best income group among whites with a college degree. Among those making between $30,000 and $99,999, the difference is about 40 percentage points.

Third, Trump saw little difference in his support between income levels within each education group. He won every single income group among those with no college degree by between 32 percentage points and 49 percentage points. He saw a 16 percentage point gap in his margin between his worst and best income group among whites with at least a college degree. Further, there isn’t a clear effect of income across most education groups. Trump’s two worst groups among college whites were those who earn less than $50,000 and those earning more than $200,000. Among whites without a college degree, one of his best groups was those earning $200,000 or more per year.