Donald Trump and the "N-word": Many voters would still support him — but not enough

Data journalist George Elliott Morris says most Republicans don't mind racial slurs, but a tape would hurt Trump

By Chauncey DeVega

Senior Writer

Published September 11, 2018 8:00AM (EDT)


After the white supremacist riot in Charlottesville last year that killed Heather Heyer, President Donald Trump infamously told reporters that there were "very fine people" on both sides. It's a comment that may define his presidency more than any other.

As former President Barack Obama said at a Democratic campaign rally last Friday, "It shouldn't be Democratic or Republican to say we don't target certain groups of people based on what they look like or how they pray. We are Americans. We're supposed to stand up to bullies, not follow them. We're supposed to stand up to discrimination. And we're sure as heck supposed to stand up clearly and unequivocally to Nazi sympathizers. How hard can that be, saying that Nazis are bad?"

Given Trump's evident embrace of white racism and his unabashed use of it to win the 2016 election, his refusal to condemn white supremacists and other hate mongers should not be a surprise. In fact, Bob Woodward reveals in his new book "Fear: Trump in the White House" that after Trump's aides and advisers persuaded him to condemn neo-Nazis and other white supremacists more forcefully following the Charlottesville riot, Trump immediately regretted it, saying that second statement was the "biggest fucking mistake" he had made.

Woodward also reports that Trump told his advisers, “I didn’t do anything wrong in the first place" and lamented that he had been “forced” to “apologize.” According to Woodward's account, Trump said, “I’m never going to do anything like that again.” Perhaps he simply meant he would never again allow himself to be put on the defensive; just as likely, he also meant he would never again condemn neo-Nazis, Ku Klux Klan members and other white supremacists.

This is all part of a larger pattern in which Trump and the Republican Party have nurtured and embraced a politics of white identity and white backlash that considers public norms or standards that reject overt racism as being at best quaint, and at worst something to be jettisoned entirely.

Are there any limits to Donald Trump's embrace of anti-black and anti-brown racism? What if the rumored tape (which his former aide Omarosa claims to have heard) of Trump using racial slurs against black people was made public? Would this damage his popularity or hurt his chances for re-election?

A new online public opinion survey conducted by YouGov and the Economist attempts to answer these questions. Data journalist George Elliott Morris explains some of the key findings:

The results confirm that most Trump voters would be unfazed: 77% of white Trump voters agreed that “it is possible that a person who uses the ‘N-word’ while in office can still be a good President of the United States”. Just 11% of whites who voted for Hillary Clinton said the same. By YouGov’s estimates, 47% of white Trump voters say that they could “definitely” or “probably” “support a presidential candidate whom you knew for a fact uses the ‘N-word’ to refer to a black/African American person”. Only 4% of white Clinton voters said the same, compared to 22% of American adults as a whole.

I recently spoke to Morris about this new poll, its implications for Trump's presidency, the shortcomings in much of the reporting about Trump's voters and their racial attitudes, and how racism (including tolerance for using anti-black racial slurs) indicates a significant divide between white Republicans and white Democrats. Morris also shares his predictions for the upcoming midterm elections as well as Trump's chances of victory in 2020.

This conversation has been edited for clarity and length.

From your perspective as an expert in public opinion, what do journalists as well as the general public continue not to understand about Donald Trump and his movement?

If I didn't understand a great deal about how decisions are made in newsrooms, I would say that journalists got it wrong with Trump in 2016, and they continue to get some of the angles of Trump's politics and maneuvering wrong at present. The public still doesn't understand the dangers that authoritarianism plays to our political institutions. But if you look at the polling data the country is at least seven or eight percentage points more Democratic. Historically this is a pretty large swing. They are feeling anxiety and concern about the Republicans. The public may not understand the more academic details of why Trump is so dangerous, but ultimately they are getting it right.

There have been rumors for some time -- coming to a head with his protégé and former White House aide Omarosa Manigault Newman -- that there are actual audio or video recordings of Donald Trump using racial slurs about black people. What does your new poll reveal about how Trump and other voters would respond if this were publicly verified?

One of the things that complicates the analysis is that nobody knows for sure whether or not Trump did or did not say these things, except for the people who were in the room with him. But what we do know is that roughly 47 percent of white Trump voters would still vote for him, even if they knew for certain he had used the "N-word." Anywhere from 70 to 80 percent of white Trump voters think that a president who said the "N-word" could still be a “good president."

In general, we know that there is a high tolerance for racism among the white Trump voters in the country. And that should not be surprising. During the campaign, Trump made many  overt racial appeals during. He kicked off his campaign by calling Mexicans, "rapists" and "criminals." Polling and other research has repeatedly shown that racial resentment is explicitly tied to supporting Trump, all other demographic factors being equal.

Consequently we found in our survey that if you're a white college-educated voter who voted for Hillary Clinton you are about 50 percent less likely to use the "N-word" than if you voted for Donald Trump. Clearly there is something about being a Trump voter that makes you more likely to use that racial slur or to accept someone else saying that word.

How can we put this in context relative to what we know about the Republican Party and its supporters more generally?

There is a portion of the Republican Party which is nativist. The data is clear on that point. There is also  a large swath of the Republican Party who support Trump and are fine with his attacks on African-Americans and immigrants. You do not see that with the Democratic Party, because it is a coalition which welcomes different racial groups. This is why the voters who are fine with using racial slurs have found a home -- to say the least -- with Donald Trump and the Republican Party.

Your new research also seems to subvert the dominant and incorrect narrative that Donald Trump's support was and remains driven by "economic anxiety" among white voters.

Let's dig into the statistics and how we as pollsters think about this puzzle. If you take only those who we know for a fact voted, and you take them at their word that they voted one way or another, and then you use all of their demographic factors to predict whether or not they voted for Trump, you actually do get a statistically significant relationship with income. So there is some element of economic anxiety in the country which obviously relates to voting and politics more generally. But what a lot of people miss is that if you control for other variables there is a very strong and statistically significant relationship between voting for Trump and racial animus against nonwhites.

As a practical matter, it matters whether a voter believes it is OK for the president of the United States to use the "N-word," or believes that immigrants should be deported or that nonwhites are taking jobs away from white Americans. If there is economic and racial anxiety in the country the data suggests that it is racial animus and racial attitudes that pushes a person to one party or the other, more so than economic anxiety and the amount of money you make.

There is a whole genre of writing by reporters and others where they go out to "Trump country" and interview his voters and other supporters at rallies, in focus groups, in their homes and the like. These are anecdotes which only seem to support incorrect narratives about Trump's ascendance to power. What does the data actually reveal about macro-level social dynamics and politics in America?

That type of journalism is not really representative of who the Trump supporters are. A reporter can go to a rally and find a Trump supporter but what they're going to tell you is probably not going to be a profile of America at large. Those types of profiles are also not going to reveal much that we don't know already: White men specifically, and white people in general, are feeling a type of status threat from women and minorities. This is why those voters are sticking with Trump and the Republican Party. They feel that Trump is the person to stave off this imagined decline in group power and status threat from nonwhites, women and other groups. Of course this sense of status threat and anxiety ignores the fact that white men are still by far the dominant group in American society.

If you are a Democratic strategist, should you view Trump's voters who have such a deep level of racial animus as being persuadable to your side? And if you're a Republican strategist seeing these findings about racism among your voters, is the takeaway that there is no electoral penalty -- or maybe there's even an incentive -- for your candidates being overtly racist?

I think that is mostly correct. Most Republicans aren't ever going to vote for a Democrat and most Democrats aren't going to vote for a Republican. This is roughly symmetrical. This polarization is especially true when we are talking about racial attitudes and how they impact party choice. When you factor in other social issues such as women's rights to reproductive freedom, the country is highly polarized and people are not going to change their minds. There is a deep social motivation for staying with whatever party you have decided is the best embodiment of your values or policy positions. This is why pollsters have come to the conclusion that if you want to win an election you have to turn out all the people who are persuaded to your side but aren't persuadable to choosing either the Republican or Democratic Party.

Turning out your own voters and supporters is what matters. When we factor in racism and racial animus in how voters make political decisions, there are not a lot of people who are going to switch sides. This is even more true now that it was two years ago. If the current trajectory in the country stays the same, the power of racism and racial animus over how people decide to vote and what party to join will become even greater.

If you were to ask Trump voters if either he or they are racists, they would be offended and angry. Like Trump himself, they would likely say they are the "least racist" people you will ever meet. What does your new poll reveal about people's private thoughts and behavior, versus what they would admit to pollsters or other researchers?

This is the advantage of using the online YouGov survey. This circumvents the social desirability effect. We let them just click a box online. This also speaks to the journalism where reporters speak to Trump's voters in person. They are usually not going to tell you that they are racists. If they have unpopular feelings, they likely will not share them with the interviewer.

What was really interesting is that Democrats and Republicans were equally as likely to tell us that they did not want to say whether they had used the "N-word" or not. This is odd, given that one would expect most people to deny using such language. If you assume those respondents have used that racial slur, as well the share of people in the survey who admitted they have said the word, it basically increases by 4 or 5 percent the number of whites in the survey who have evidently used the "N-word."

READ MORE: Another ho-hum summer with Donald Trump: All the season's big news stories you wish you could forget

One of the most interesting findings of your new public opinion research was that 20 percent of Hillary Clinton voters admitted to using racial slurs for black people. Of course, that number was much higher for Trump voters.

Thirty-seven percent of Trump voters admitted that they had said the "N-word," while 20 percent of Hillary Clinton voters admitted the same thing. It is not a zero-sum game where if you are a racist you are with Trump and if you're not, you're not. There are people who hold such attitudes among both Hillary and Trump supporters. But only 4 or 5 percent of Hillary voters report being OK with a president who used that racial slur for black people, versus 77 percent of Republicans.

Let us imagine that a video or audio tape emerges where Donald Trump clearly and undeniably uses racial slurs against black people or other nonwhites. Given how polarized the country is, and the way the Republican Party cultivates and encourages racism, would that even hurt him? 

Our survey finds that 47 percent of white Trump voters would vote for someone again who used the "N-word." But that also means that 53 percent of them either wouldn't vote for a person who used such a slur, or don't know whether they would. That is a lot of people, which could have a dramatic effect on American politics from the presidency to Congress on down. If our data is correct, Donald Trump would probably lose the presidency if a tape of him using racial slurs were to be released. But on the other hand, there are many voters who would still support him.

Of course, this is all part of a much bigger story. What do you think will happen with the midterms? What are Trump's chances for re-election? 

The Democrats have somewhere between a 70 and 80 percent likelihood of taking the House of Representatives, if the election was held today. That might increase or decrease as we get closer to November. I would put my money on that number increasing. The number is definitely lower than half for taking the Senate. As far as Trump being elected in 2020 goes, I'm hesitant to come to a firm conclusion so early. However, if Donald Trump's approval rating is at 40 percent on Election Day 2020, and if the economy is where it is or better than today, then Trump probably has pretty OK chances of being re-elected.

But if there's a recession or his approval is between 35 and 40 percent -- which is kind of hard to imagine, given the stability of Trump's support among Republicans and his other supporters -- then he might be in danger.

Can the Republican Party survive Donald Trump?

Rick Wilson tells Salon Talks how the GOP can find its way back after the Trump takeover

By Chauncey DeVega

Chauncey DeVega is a senior politics writer for Salon. His essays can also be found at He also hosts a weekly podcast, The Chauncey DeVega Show. Chauncey can be followed on Twitter and Facebook.

MORE FROM Chauncey DeVega