Who are Trump Voters? Social Sciencing the S**t out of Yard Signs

Last week I blogged about how there are no Trump campaign signs to be found anywhere inside the Beltway, and naturally I saw one the next day. But the absence of any others in a town full of Grand Old Party operatives, defense contractors and the like has been a real puzzle for me, especially since driving out well beyond Washington yields an unbroken sea of Trump signs in parts of Maryland and Pennsylvania, two states Clinton is in no danger of losing.

The question has become vexing enough for me that I needed to put some real work into figuring this out. It was time to turn to my training as a social scientist, as Emile Durkheim and Robert Putnam had done previously, to rule out spurious correlations and identify the variable that best explains the phenomenon. In _The Martian, _Matt Damon’s biologist declared that he was going to survive being stranded on the red planet by using ‘science’ as a verb (‘I’m going to science the shit out of Mars!’) I decided to explain red voters (the colour typically used for the GOP on electoral maps) by social sciencing the shit out of Maryland. So here goes: _

The one and only DC area Trump sign I’ve seen in six weeks back here was a small red ‘Maryland for Trump’ (as opposed to the standard blue Trump logo) in front of a house on River Road in Bethesda, just a few blocks away from the Capital Beltway onramp. Obviously a highly aspirational local party activist lived there: Although a previously unheralded Republican businessman does currently serve as governor of Maryland, the Free State is one of the most Democratic in the country. The presidential and open senate race are no contests this year, and the US House delegation is easily expected to remain 7-1 Democratic.

But drive an hour or so east of Washington into that one Republican district and Trump signs cover cars, farms, houses and billboards. The same is true if you continue north across the old Mason Dixon line (the boundary of slavery before the Civil War) into Pennsylvania. Signs for Trump and local Republican candidates abound (although there were strangely almost none for incumbent senator Pat Toomey, who is in danger of losing and costing his party control of the Senate) with no Clinton signs visible whatsoever.

So what gives? Who is proudly announcing that they are voting for Trump and who is not?

There have been many predictions and explanations about likely Trump voters this year, but most of them don’t appear to fit my observations. Some pundits have speculated that Trump would attract legions of economically and culturally disaffected blue collar white voters, essentially recreating the Nixon and Reagan coalitions. Others have noted that Trump’s supporters appear to have an above-average household income of $72,000, so he’s actually appealing to well-off white Republican-leaning voters who expect tax breaks and want to cut the social services that they believe disproportionately benefit poor minorities. (In fact, whites receive the biggest share of welfare benefits). Here’s what the demographics show:

Eastern shore Maryland and southern Pennsylvania (outside of the Philadelphia area) are Republican strongholds in presidential and congressional elections. It’s no surprise to find Trump support there. And despite their rustic reputations as havens for Chesapeake Bay crabbers and Amish farmers, the median income in these counties runs over $65,000, above the national average but in average Trump supporter territory.

But if money is the answer then why is only one humble Trump sign visible in Montgomery County, Maryland, with a median income of $95,000? Trump signs abound among well-off Marylanders and Pennsylvanians on both signs of the Susquehanna River, but not among far richer Marylanders and Virginians along the Potomac River just outside of Washington DC. Potomac, Maryland and McLean, Virginia can regularly claim to be two of the top three wealthiest suburbs in America. Although partisan redistricting gave the former a Democratic representative in Congress last decade, Dick Cheney lives in a long-time Republican district in the latter. And yet there are Clinton and even still Sanders signs in abundance but none for the GOP presidential nominee.

It is doubly strange because these wealthy enclaves tend to skew older (not too many first time home buyers). Since Franklin D. Roosevelt created Social Security in the 1930’s, retirees have been the most well-off demographic in America. And because Trump is doing best amongst the richest and the oldest voters, he ought to be over-performing in older, richer communities with baked-in clusters of Republican Party leaders and donors.

If partisan affiliation, age, and income levels are not sufficient to explain the absence of Trump support, what is? This social scientist is left with another variable as the best explanation: Education levels. Trump is performing strongest amongst those without higher education and is historically weak for a Republican presidential nominee amongst those who have been to university. Washington DC ranked this year as the #1 or #2 most educated American city (depending on the index). The pro-Trump counties outside the beltway I have described (Queen Anne’s, Kent, and Chester in Maryland, Chester and Lancaster in Pennsylvania) have higher than average household income levels, but lower than average rates of bachelor degree completion).

Now I realize I’ve just provided plenty of fodder for the comments section. To take a line from another film, Ghostbusters, ‘Back off man, I’m a [social] scientist!’

I know that this will not stop the trolls on both sides, but I am not making the argument that Trump supporters are ignorant. I am just noting that the best correlation I can find between displays of Trump support in the parts of three states and the District of Columbia I’ve driven through in August and September 2016 seems to be levels of higher education rather than party or income.

As Sherlock Holmes said, ‘Once you eliminate the impossible, whatever remains, no matter how improbable, must be the truth.’ I’m not claiming ‘truth’ here – to quote yet a third film, as Indiana Jones told his students in this third outing, ‘Archaeology [substitute Political Science] is the search for facts, not truth.’

In the absence of any better data, I’m declaring case closed. Now go watch the debate.