One of many analyses to come (I hope)
A British study seems to confirm parts of conventional wisdom, but leaves unresolved two very unpleasant questions.
The Resolution Foundation — a British “think-tank” has conducted an analysis of the U.S. presidential election results at the county level. I highly recommend reading it. The method looks at changes in voting patterns — rather than the absolute share — and runs regression analyses against several individual analyses, and then against the collection of them.
The analysis shows that the share of white voters with only a high school education was the largest single determinant of shifts in Trump’s favor. Geography, economic conditions and outlook also had a significant impact, but faded compared to education and “race.”
This is likely because — as the analysis shows — there is a high correlation between educational attainment and household income.
Still, non-white voters with similar economic and educational level were more likely to vote for Clinton. In other words, all else being equal, being white and male made the biggest difference. Even if, perhaps especially if, you voted for Obama four years ago.
All this does not explain what motivated white voters without college degrees, and especially of the male variety, to vote for Trump — and in my view — against their own interests. More research is needed.
This research is likely to be unpleasant for everyone. It should make Democrats uncomfortable that real incomes for this segment are stagnant while costs for education and health care have easily outstripped inflation. Republicans should be ashamed at having encouraged racism and misogyny as a reaction to this injustice.