You have to hand one thing to America’s conservative white evangelicals. Over the last few decades, they’ve become experts at marketing, PR, and image management, using vehicles like the National Prayer Breakfast and The Gathering to slickly handle relationships with political elites and the press.

Major media players like the New York Times and Washington Post sense that they might be susceptible to charges of bias, particularly when such charges challenge the objectivity of editors and reporters, many of whom are urban liberals. Evangelicals have been quite critical of the media for failing to understand them. After the 2016 election the Times‘ executive editor told NPR: “I think that the New York-based and Washington-based too probably, media powerhouses don’t quite get religion.”

The problem is that efforts to better understand evangelicals all too often manifest as a reluctance to be critical and a failure to include the perspectives of both exvangelicals and policy researchers, both of whom are knowledgeable and have legitimate concerns about widespread evangelical tendencies toward the rejection of pluralism and attempts to impose theocracy. In 2014, according to a Christian Post report, an evangelical Times editor, Michael Luo, “told an audience at The King’s College… that whenever his colleagues at the New York Times do happen to botch a report that casts Evangelical Christians in an unfair light, that it is mostly due to ignorance.” Although charges of unfairness are often spurious and a manifestation of conservative evangelicals’ persecution complex, it’s clear why editors and reporters might be reluctant to bring their full critical faculties to bear on evangelical subculture, particularly given how alien it is to many of them.

As a result, in order to avoid kerfuffles, major media outlets often have reporters who are current or former evangelicals, or who are extremely friendly to evangelicals, do the lions’ share of the coverage of evangelicalism. In the Washington Post, evangelicals largely cover themselves. Both Michael Gerson and Sarah Pulliam Bailey are graduates of Wheaton College, arguably the epicenter of “respectable” evangelicalism. And a handful of evangelicals who have become experts at presenting a moderate face to the media are far and away the most frequently quoted sources. Their answers to softball questions are left unchallenged, and there is little or no probing into those areas in which exvangelicals and policy researchers would easily see answers as incomplete or outright disingenuous.

For example, Russell Moore, president of the Southern Baptist Convention’s Ethics and Religious Liberty Commission, is quoted constantly, but no journalist on the staff of any major outlet seems interested in asking him whether he still considers pluralism and feminism “serious problems in Evangelical Christianity today” and “outright heresy,” as he wrote in 2007. Yet if we’re interested in serious journalistic coverage of conservative evangelicals rather than what all too often amounts to free PR, journalists will need to ask such questions, to focus on the ways that evangelicals make themselves “all things to all people,” addressing the general public very differently from the ways they address their own; and journalists will need to include the perspectives of critical researchers and exvangelicals alongside the voices of evangelicals in the articles they produce.

Because we have very few such journalists—Nina Burleigh of Newsweek is an exception—the U.S. media has contributed mightily to the normalization of white evangelicals, an overwhelmingly anti-pluralist, anti-democratic demographic that researchers have identified as “uniquely conservative.” However, thanks to white evangelicals’ unwavering Trump support—81 percent voted for him and 71 percent still view Donald Trump’s presidency favorably—we’ve recently seen an unusual amount of critical coverage in typically friendly outlets over white evangelicals’ prevailing attitudes toward asylum seekers and refugees. Evangelicals—and those who are strongly invested in a “respectable” image for evangelicalism—now seem to be making a concerted effort to push back ahead of the midterm elections, in which evangelicals’ reputation will be tied to what many will see as a referendum on a historically unpopular authoritarian president for whom they remain the single most supportive demographic.

The first salvo in this new PR blitz was launched in September, when the Democracy Fund’s Voter Study Group published a report, written by the director of polling at the libertarian Cato Institute, that purports to show that “Trump voters who attend church regularly are significantly more likely than nonreligious Trump voters to have favorable attitudes toward black people, Hispanics, Asians, Jews, Muslims, and immigrants, even while holding other demographic factors, such as education, constant.” In a New York Times op-ed, the report’s author Emily Ekins framed the report explicitly as a response to “people on the left” who “think the religious right has compromised its Christian values in order to attain political power for Republicans.” The problem with Ekins’ claim that “church teachings can curb tribalistic impulses by regularly reminding worshipers that we are all God’s children,” is that her own data, which she is employing either sloppily or disingenuously, shows nothing of the sort.

In a thorough analysis for RD, Timothy Gloege points to findings in the Voter Study Group data that make the uncritical acceptance of Ekins’ claims impossible for any intellectually honest observer. For example, the study found that 82 percent of frequent church-attending Trump voters agree with the statement that “if blacks would only try harder they could be just as well off as whites,” and a whopping 91 percent agree that “Irish, Italian, Jewish, and many other minorities overcame prejudice and worked their way up. Blacks should do the same without any special favors.” What we have here are “respectable” racists, the kind who are image-conscious enough not to want to be seen as racist, deflecting such charges with declarations that, “I can’t be racist, I have black friends.” What we certainly do not have is evidence that churchgoing Trump voters are in actual fact less racist than secular Trump voters.

In the Washington Post, Paul A. Djupe and Ryan P. Burge make much the same point, drawing attention to the fact that frequent church attendance correlates with higher rates of Trump support. In addition, the data shows that for Trump voters who never attend church, self-reported positive feelings toward minorities correlate with less support for the president, but among churchgoing Trump voters the opposite holds true. So what’s going on? According to Djupe and Burge, “We suspect that people are responding to these surveys with the responses they think are more socially desirable—not with their actual feelings.” The fact that people may lie in answering survey questions is fundamental to doing serious sociological research. The evidence for this, which Ekins conveniently ignores, is overwhelmingly present in her own data.

But the Cato Institute isn’t doing all the heavy lifting when it comes to “saving” evangelicals’ reputation. Evangelical leaders and institutions have also been working to distance themselves from Trump in the public’s perception. Just as conservative evangelicals cherry-pick their favored Bible verses in order to focus on punitive readings obsessed with policing sexuality, they also have a knack for cherry-picking data that paints them in a positive light.

On October 18, Ed Stetzer and Andrew McDonald published a write-up of a study in Christianity Today that uses data from Wheaton’s Billy Graham Center Institute (of which Stetzer is executive director) and Lifeway Research (a branch of a Southern Baptist publishing and retail juggernaut). Their analysis focuses on evangelicals’ self-described motives for voting for Trump, as if people don’t lie about their motivations in surveys such as this, and ignoring the possibility that voting for and continuing to support a president who separates the children of asylum seekers from their parents at the border and who stands accused of sexual misconduct by 22 women is somehow morally justifiable if one claims to do so for economic reasons, or to have simply voted “against Hillary Clinton.” Stetzer and McDonald also awkwardly try to argue that abortion and the Supreme Court were not key reasons for white evangelical Trump support, despite noting that abortion is a critical reason many white evangelicals categorically refuse to vote for Democrats.

The Billy Graham Center/LifeWay study attempts to get results similar to Ekins’ on race by differentiating between self-identified white evangelicals and “evangelicals by belief.” This seems reasonable enough until it becomes clear that the primary function of the “evangelicals by belief” category is to conflate Christians of color (who may not even consider themselves evangelicals) with white evangelicals on the basis of certain shared theological views, in order to make white evangelicals come off as more moderate than they are. Daniel José Camacho has provided a thorough takedown of the Christianity Today article at Sojourners, noting that “The study’s findings are based on a colorblind lumping of all evangelicals according to belief [which] actually makes it harder to determine and compare what motivated white evangelicals as opposed to evangelicals of color in the 2016 presidential election.”

Once again, we have a disingenuous deflection from the very real issue of white evangelical racism that mirrors the “but I have black friends” defense. As Camacho says, “The study argues that it includes overlooked evangelicals of color, but it includes them as pawns that shore up the reputation of white evangelicals.” One can argue that this has also been the thrust of decades of efforts that white evangelical leaders like Russell Moore have poured into so-called “racial reconciliation,” a project that has arguably failed, as we are seeing Black people and other people of color leave predominantly white evangelical churches when their concerns about the racism and Trump support of their fellow church members are not taken seriously.

Indeed, one can even argue that white evangelicals’ abortion obsession is a “respectable” stand-in for the racial animus at the root of the contemporary Christian Right. What an intellectually honest observer cannot do, on the basis of the findings written up by Stetzer and McDonald, is pretend that abortion and race are unimportant matters when it comes to white evangelical voting. And even if we could, to support this president despite his continuous assaults on the free press, his overtures to white nationalists, his human rights violations at the border, and his anti-woman and anti-LGBTQ policies, is morally despicable regardless of motive.

People may lie to themselves as well as to pollsters about their own motives. Facing complicity in such enormous cruelty represents a serious ego threat, but if evangelicals want to be taken seriously by the U.S. mainstream going forward, they are going to need to prove themselves capable of serious self-reflection and repentance. To give credit where credit is due, Stetzer has argued in the Washington Post  that “God’s people should be the first ones to open their arms to refugees.” Nevertheless, so long as Stetzer is at least as dedicated to saving evangelicalism’s image as he is to calling for evangelicals to be better, he remains complicit and part of the problem. White evangelicals own the disastrous Trump presidency. Until they admit this, admit it is a problem, and change, we must recognize them as a toxic illiberal force in American society and politics, one that needs to be politically sidelined if we are going to preserve American democracy.