Something to file away. Bitzer is a professor in North Carolina and was a useful Twitter resource during both the South Carolina and North Carolina primaries.

Running some #s on Trump’s results in North Carolina’s 100 counties & what county-level factors that might explain his performance #ncpol

— Dr. Michael Bitzer (@CatawbaPolitics) March 20, 2016

First, % of black registered voters in county to Trump’s performance (possible racial resentment?). Not much there pic.twitter.com/g2jLA4L9SD

— Dr. Michael Bitzer (@CatawbaPolitics) March 20, 2016

Next, % of county’s registered voters who are Baby Boomers & Silent Gen (age factor?). More, but still meh. pic.twitter.com/COCeIqcXHH

— Dr. Michael Bitzer (@CatawbaPolitics) March 20, 2016

Next, county’s Jan 2016 unemployment rate (economics factor?) to Trump vote. Little bit more there than age pic.twitter.com/2mfkas1wGx

— Dr. Michael Bitzer (@CatawbaPolitics) March 20, 2016

Most notable factor so far to Trump NC vote: county level of bachelor degree or higher (2009-13), but very negative pic.twitter.com/th8W1e0HMl

— Dr. Michael Bitzer (@CatawbaPolitics) March 20, 2016

Here is the full exit poll, which makes for useful comparison. It has varied from state to state, but it is definitely the case in North Carolina that the more educated a voter is, the less likely they are to vote for Trump. Even with that, though, it should be noted that he won 35% of college graduates and 30% of postgrads. In terms of income, he won mostly on the basis of those making $30-50k a year, though again did impressively across the board (and won among those who make more than $100k a year).