January 18, 2018

The Role of Race in Trump’s Victory and the 2016 Election

Since Donald Trump won, there’s been a lot of talk about the role that race played in the election, so let’s take a look at that. Here is how various ethnic groups have voted in US presidential elections, including the one that just happened, based on New York Times exit poll data:

Here is how the democrat advantage, or disadvantage, among various ethnic groups has changed with time:

Dem advantage by race by year.png

Perhaps surprisingly, 2016 doesn’t pop out as special in any way in these charts. However, something interesting happens if we separate winning and losing republican presidential candidates:

How Winning and losing republicans, and Trump, did among different ethnic groups.png

Trump did better than losing republicans, and as well as, but not better than, previous winning republicans, among Whites. He also did better than losing republicans, but not as well as the typical winning republican, among Hispanics. He did worse than the typical winning or losing republican among African Americans and Asian Americans.

So yes, Trump won because of his performance among Whites (and Hispanics), but no more so than past winning Republicans.

A lot of attention has also be paid to how different kinds of White people voted. Here is a basic breakdown from CNN exit poll data:

How White Americans Voted 2016.png

That education gap has gotten a lot of attention. Let’s take a look at that over time. Here is how Whites have voted over time based on whether or not they have a college degree:

And here is a similar line of data on Whites categorized by sex and whether or not they ever went to college:


Clearly, democrats are losing non-college educated Whites in a big way and Trump capitalized on this like no republican ever has. College educated Whites, by contrast, did not vote for Trump by as large of a margin as they have voted for other republicans. Given this, while Trump may have one the same share (of two-party voter) Whites as past winning republicans have, they probably weren’t the same kind of Whites in terms of formal education.

If we look at the important swing states that Trump won, again using CNN exit polls, we see a clear pattern: compared to the national average they have an abnormally high proportion of the population which are White and which are Whites with no college degree.

Whites in these states were not, however, more pro-trump than was typical for White voters, or White non-college degree holding voters, in 2016. There were just more of them.

Key State Demographics.png

With Hispanics we see the opposite story: in these key swing states there were fewer Hispanics than average, but these Hispanics were also considerably more likely than average to vote for Trump. This trend is particularly noteworthy in Florida, Michigan, and North Carolina, where Trump won more than 35% of the Hispanic vote.

In fact, in Florida, Trump won 54% of the Cuban vote. Florida is also interesting because the Whites there were at once abnormally pro trump and more likely than average to have a college degree.

North Carolina is the only states that these demographic variables can’t explain Trump’s victory in. Frankly, I have no idea why he won North Carolina.

To sum up:

  • Like most republicans, Trump won by doing well among Whites. However, these were not the same Whites that traditionally carry republicans to victory.
  • Yes, Trump won big among Whites without college degrees, but Whites with college degrees voted for him as well. Given this, the notion that Trump voters are “hicks” or “white trash” doesn’t hold up.
  • In spite of his rhetoric on immigration, Trump did fairly average, for a Republican, among Hispanics, and better than most losing Republican candidates have done. In fact, abnormally pro-Trump Hispanic populations may have been key to him carrying several swing states.


Facebook Comments