The myth of the myth of Trump’s working-class support

Data journalism! Good lord. Trump voters have a higher median income than the population as a whole:

As compared with most Americans, Trump’s voters are better off. The median household income of a Trump voter so far in the primaries is about $72,000, based on estimates derived from exit polls and Census Bureau data. That’s lower than the $91,000 median for Kasich voters. But it’s well above the national median household income of about $56,000. It’s also higher than the median income for Hillary Clinton and Bernie Sanders supporters, which is around $61,000 for both. …

Ted Cruz voters have a similar median income to Trump supporters — about $73,000. Kasich’s supporters have a very high median income, $91,000, and it has exceeded $100,000 in several states. Rubio’s voters, not displayed in the table above, followed a similar pattern to Kasich voters, with a median income of $88,000.

In other words, every candidate’s supporters have a median household income higher than the national average. Why should we be surprised that Trump’s do? Trump’s supporters have a lower median income than any other major Republican candidate’s, and that’s taking into account that Trump is most popular on the East Coast, where the cost of living is higher.

But it gets worse. The median isn’t even relevant here. To judge whether Trump has working-class support, we should look directly at the working class. Let’s say the working class is anyone who never went to college, and look at some state-level exit polls. (This won’t be a thorough statistical analysis, and it could be that the states I don’t look at show the opposite pattern, but I’ll take all of the first states and a few of the last.)

In Iowa, Trump won 32% of voters with an education level of high school or less, and Cruz won 28%. (Trump also won 34% of moderates in Iowa. 28% went to Rubio, and no other candidate won more than 10% of them.)

In New Hampshire, Trump won every single demographic the exit poll measured, except people who think the most important quality in a candidate is sharing their values (21% Cruz, 20% Kasich, 13% Rubio, 13% Trump), people who are only somewhat worried about the US economy (27% Kasich, 25% Trump), people who oppose a temporary ban on Muslims entering the US (26% Kasich, 17% Bush, 16% Trump), people who think the next president should be experienced in politics  (27% Kasich, 21% Bush, 17% Rubio, 14% Cruz, 11% Christie, 6% Trump), people who won’t be satisfied if Trump wins the nomination (29% Kasich, 20% Bush, 17% Rubio, 14% Cruz, 10% Christie, 5% Fiorina, 3% Carson, 2% Trump), people who think Kasich would best handle the economy (who make up 18% of their sample, as opposed to 41% for Trump; no other candidate scored high enough for CNN to provide a breakdown), and people who had only decided who to vote for in the last few days (24% Kasich, 22% Trump). This looks like a lot in text, but it’s only seven of a few dozen categories; and, remember, Trump won all of the rest.

In South Carolina, Trump won 45% of voters who have never gone to college. He also won every other education demographic, except people with a postgraduate education, where he came second behind Rubio. (Among college attendees as a whole — that is, ‘some college’, ‘graduated college’, and ‘postgraduate’, combined into one category — Rubio won 27% and Trump won 25%.) This exit poll has income data; Trump won every income demographic, but tied in $100k-$200k.

In Nevada, Trump won every education demographic.

In Alabama, he tied with Rubio among people with postgraduate education and won every other demographic. The Alabama exit poll has income data, and Trump won every income demographic.

In Arkansas, Trump came third among postgraduates, behind Rubio and Cruz, but won every other income demographic. The Arkansas exit poll has income data, and Trump came second behind Cruz in $50k-$100k and won every other one.

That’s enough of the early states; now let’s look at the late ones. Maybe some other candidates were splitting the vote…

Or maybe not. In Connecticut, Pennsylvania, and Maryland, Trump won every education demographic and every income demographic.

These are all East Coast states, so maybe the low end is getting cut off. How about Missouri?

In Missouri, Trump won in the high school or less and some college demographics, and came second behind Cruz among college graduates and postgraduates. Cruz won $30k-$50k and $100k-$200k voters, and Trump won $50k-$100k voters; where Trump won, Cruz came second, and vice versa.

In Illinois, Cruz and Kasich tied among postgraduates, and Trump won every other education demographic. Only $50k-$100k and $100k-$200k reached significance for CNN here, but Trump won both.

In Wisconsin, moderates went for Trump and everyone else went for Cruz. (These Inland North people — they’re too nice. Too nice. How do they plan to make it through this century? This century won’t be nice to anyone, believe me.)


5 responses to “The myth of the myth of Trump’s working-class support

  1. haishan May 3, 2016 at 9:08 pm

    The problem with data journalism is that it’s still journalism. Journalists will twist the data to serve journalists’ ends.

    A preliminary look at the exit polling data suggests that Trump exhibits large and consistent overperformance among non-college graduates, and perhaps some degree of overperformance among low earners as well. (Not consistently, though — he actually did better among voters earning over $50,000 in Maryland, for example.) I’ll probably take a deeper look into this later.

    • nydwracu May 4, 2016 at 2:29 am

      …In fact, 538’s metric is so obviously the wrong one to use that I can’t imagine it was an honest mistake. Not that “honest mistake” should ever be your first hypothesis when you see a story in the news, but some people, even some with the proper skepticism toward the rest of the Pravda complex, are apparently under the delusion that nerd-branded outlets like 538 and Vox will ever not lie.

      (Remember Walter Duranty. They’re all Walter Duranty — some of them just have different Stalins.)

      • Alrenous May 4, 2016 at 3:48 pm

        Douthat & recent Vox Day coverage is revealing they can be not-stupid if they want to be. There used to be some doubt about whether they really were that deluded.

  2. Pingback: The myth of the myth of Trump’s working-class support | Reaction Times

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: