Op-ed by Thomas B. Edsall @ NYTimes.com, March 31

[....] “The short answer is that the exit polls are wrong,” Matthew DeBell, a senior scholar at Stanford’s Institute for Research in the Social Sciences, emailed me. He continued:

In November 2016, 31.9 percent of adult US citizens had college degrees, according to the Current Population Survey. There were 138.8 million votes. To reach 50 percent of all voters, the turnout rate among college grads would have to have been 97 percent. This doesn’t pass the laugh test; no credible study has ever found turnout rates that high.

The Pew Research Center and the Center for American Progress have produced methodologically sophisticated surveys of the electorate that sharply contradict 2016 exit polls.

Perhaps most significant, a March 20 Pew Research Center public opinion survey found that 33 percent of Democratic voters and Democratic leaners are whites without college degrees. That’s substantially larger than the 26 percent of Democrats who are whites with college degrees — the group that many analysts had come to believe was the dominant constituency in the party.

According to Pew, this noncollege white 33 percent makes up a larger bloc of the party’s voters than the 28 percent made up of racial and ethnic minorities without degrees. It is also larger than the 12 percent of Democratic voters made up of racial and ethnic minorities with college degrees.

In sum, Pew’s more precise survey methods reveal that when Democrats are broken down by education, race and ethnicity, the white working class is the largest bloc of Democratic voters and substantially larger than the bloc of white college-educated Democratic voters [....]