Trump and white evangelical voters

I don't envy the choices that US citizens had during their recent presidential elections. Were I eligible I would not have wanted to vote for either of the main candidates, and even the Libertarian candidate does not seem that libertarian. In the lead up to the election and the aftermath there has been much analysis but I have not found it convincing. A  recent piece in the Atlantic by Merritt doesn't do much better.

One of the few things he gets correct is the appeal not to leave evangelicalism because of disillusionment over the election. There comes a time to leave groups but he is correct that this is not the time to be walking away from evangelicalism—not least because the US is smaller than the rest of the world.

But I want to challenge several of the assumptions here because the issue is a little more complicated than: lots of evangelicals voted for Trump.

81 percent of white evangelicals voted for the Trump ticket—a higher percentage than voted for George W. Bush, John McCain, or Mitt Romney.

A major premise of the article is that white evangelicals support Trump and in higher numbers than other Republican candidates of earlier elections. But did they? And what else explains their voting behaviour?

Using Merritt's source (which is exit polling data and will have a margin of error) we see that the numbers of white evangelicals for Republican and Democrat are

Percent white evangelicals voting for Republicans and Democrats
2004 R78% D21%
2008 R74% D24%
2012 R78% D21%
2016 R81% D16%

It is uncertain whether there is anything in such a small percentage change. It may well be within the margin of error. What it clearly does tell us is that historically evangelical whites have voted Republican and continued to do so this election consistent with historic trends.

But percentages do not tell us how many people voted, they tell us proportionally how people voted. If everyone voted the percentages would be helpful but not if the turnout is significantly different.

This election an estimated 133 million votes were cast* (55% of eligible voters), 129 million for Trump or Clinton. In 2012 129 million votes (55%). In 2008 131 million votes (58%). In 2004 122 million votes (57%).

Votes (millions)
2004 Bush 62 Kerry 59
2008 Obama 69 McCain 60
2012 Obama 66 Romney 61
2016 Trump 63 Clinton 66

But combining these tables is quite hard. We need to know how many white evangelicals actually voted. Pew tells us that the electorate is composed of 26% white evangelicals which is essentially unchanged since 2008. Do they mean by electorate voters or potential voters? If the later (which is what electorate usually means) that isn't definitive because potential voters are not actual voters and it could be that white evangelicals disproportionately vote (or don't vote) or didn't vote as much in a specific election. Either way, what we actually need to know is how many white evangelicals voted for Trump (or Clinton) and what that number is as a percentage of eligible white evangelical voters (each election), and has that changed over the last couple of decades.

Therefore we do not have definitive data to say that white evangelicals voted for Trump more enthusiastically than Republican candidates of yesteryear, nor can we say that the voting behaviour this year was significantly different based on reasonable inferences from data we do have. We see the same old same old, just as black females voted Democrat like they have done for the last several decades.

Which brings us to the second issue that Merritt fails to mention: Trump's rival. Merritt quotes several evangelicals who were unhappy with Trump as a candidate, to which I concur. (Ironically, that he can name so many examples kind of works against his argument). A large number of my conservative friends and colleagues disapproved of Trump. I have never heard more negative comments spoken by conservative Christians about a politician running on a conservative ticket. But one can't discuss Trump in isolation. He was running against Clinton. And a large number of conservatives have concerns about her also. They find her as dishonest as Trump if not more so. Interestingly—contra my comment about Trump above—only one of my liberal Christian friends denounced her. Now one need not necessarily vote for Trump for fear of a Clinton presidency, but it is an understandable position. Who did Merritt vote for? He did not say, though obviously not Trump. But Merritt's argument works both ways. If Christians want to wash their hands of evangelicalism because some voted Trump, cannot other Christians want to wash their hands of those who would vote for Clinton?

---

*Accurate numbers are hard to come by, this is a low estimate.