Evangelicals are not flocking to Trump--at least thus far

Pundits and reporters continue to assert that evangelical Christians are backing Donald Trump for president. For example, R.R. Reno, the editor of First Things, writes in Saturday’s Washington Post: “Donald Trump won the white evangelical Christian vote in the South Carolina primary and the Nevada caucuses.” He states as fact something not supported by available data—namely that from primary and caucus exit/entrance polls.

Further, as J.D. Vance notes in National Review, the religious-identification question asked by the Edison Research polls isn’t likely to yield reliable data anyway. We don’t know the actual question asked, but the result is listed under the heading “White evangelical or white born-again Christians.” Apparently, pollsters are asking people to self-identify (“yes” or “no”) when presented with that description (presumably, pollsters leave out the “white” and judge that by appearance).

Vance states the problem with this approach, with particular reference to South Carolina:

“The South Carolina election results suggest that practicing Christians in the state voted differently than their peers who attend church less regularly. Take, for instance, one of Trump’s strongholds, the area in and around Barnwell County, near the central part of the state. Trump won nearly 43 percent of Barnwell County, while [Ted] Cruz collected less than 20 percent. Unsurprisingly, church-attendance rates in Barnwell lag behind those in the rest of the state. Compare that to Greenville County, which has one of the highest rates of church attendance in the state: It was one of Trump’s worst counties. The pattern generally holds across South Carolina: Cruz does well where people regularly go to church; Trump does better where they don’t. The so-called Evangelical split is just a mirage, a consequence of a country (and a state) that mostly self-identifies as Christian but manages to largely avoid the pews.” (Emphasis added.)

Still, even if one takes the “White evangelical or white born-again Christians” numbers at face value (not advisable for the reasons noted above), Trump received just 21 percent of this vote in Iowa, 28 percent in New Hampshire, 40 percent in Nevada, and 34 percent in South Carolina. In other words, he has averaged about 30 percent among self-identified evangelical/born-again Christians. Or put another way, seven out of 10 self-identified evangelical/born-again voters have (thus far) not voted for Donald Trump.

There is, of course, no way to determine how many self-identified evangelical/born-again voters are committed believers. If we assume—generously perhaps—that two-thirds of those who self-identify are “the real deal,” that would put Trump’s evangelical support in the 20 percent range at best.

A few years ago, I heard sociologist Brad Wright (author of Christians Are Hate-Filled Hypocrites … and Other Lies You’ve Been Told) suggest an “experience” test when presented with polling data: “Do the data (or the conclusions drawn from the data) match my experience?” Obviously, one’s experience can be quite narrow and therefore not a good gauge. But it is worth asking in the face of an assertion such as “Trump has strong support among evangelicals” if that matches your own experience.

If the data (or conclusions from it) seem at variance with your personal experience and observation, it may be a sign that the data (or conclusions) are flawed.