Surveys about “media trust” suffer from a definitional problem. “Do you trust the media?” is a meaningful question only if we know what “the media” is. Is it The New York Times and CNN? Fox News and Breitbart? Occupy Democrats and your uncle’s memes on Facebook?

In Gallup’s data on that question — which asks about “the mass media, such as newspapers, TV, and radio” — 72 percent of Americans trusted the media in 1976, post-Watergate. By 2016, that was down to 32 percent. But the media in 1976 was your local daily, Walter Cronkite, John Chancellor, and Harry Reasoner. “The media” is something fundamentally different now, and a decline in trust is a rational reaction to that, even in an environment less polarized than our own.

All this is to say that I find trust questions about specific news organizations a bit more useful, since you know with much greater confidence what the person being queried has in mind. And we have a new data set looking at just that, from Simmons Research.

Simmons surveyed 2,009 Americans in August asking them whether or not they trusted 38 different news organizations. Here are the results:

A few things pop out here. The bottom six — Breitbart, DailyKos, Palmer Report, Occupy Democrats, InfoWars, and in last place, Daily Caller — are all explicitly partisan sites without a pre-Internet legacy brand to fall back on. (Three liberal, three conservative.) I’d argue there are important distinctions to be made between them — InfoWars’ status as our nation’s leading lizard-people conspiracy den makes it stand out — but the average American seems to lump them together as untrustworthy.

At the top sits The Wall Street Journal, whose combination of respected news pages and conservative editorial pages seem to be a magic formula for generating trust across the ideological spectrum. (When Pew Research did a similar exercise in 2014, the Journal was the only publisher to be more trusted than untrusted in all ideological subgroups.) Then come the major networks, then the national newspapers, then wire services. A few surprises, to me: Forbes and The Washington Times ranking as highly as they did, and Mother Jones and Slate ranking as poorly as they did.

Fox News finished roughly in the middle, behind all the other cable and network news operations, but still with 44.7 percent saying they trust it. (In completely unrelated news, Donald Trump got 46.1 percent of the vote in 2016.) Fox finished tied with The Economist.

“The Doubters”

Simmons highlights one depressing finding: 13 percent of Americans said they found none of these news outlets trustworthy. Simmons terms this group “the Doubters.” (Passing up the equally applicable “the LOL Nothing Matters.”)

This group of info-nihilists is less likely to vote than the median American, but Simmons nonetheless estimates they made up about 6 million of the 129 million who cast votes in 2016. And they were much more likely to support Trump (62.2 percent) than Hillary Clinton (27.8 percent).

(This is my math, not Simmons’, but if those percentages are right, the Doubters netted out to contribute a 1.464 million vote margin for Trump. That’s about 1.13 percent of all votes cast. Trump won Michigan, Pennsylvania, and Wisconsin by 0.22, 0.72, and 0.76 percent, respectively.)

Who are these Doubters? Simmons asked these questions in the context of a larger consumer survey, so they have some interesting detail. Doubters are:

Much less likely [than the average American] to be a Democrat, much more likely to not identify with any political party at all, but just as likely to be Republican as anyone else

More likely to have married at some point, but also more likely to be widowed/divorced

Much less likely to be well educated

Less affluent

Less likely to be politically knowledgeable

Less likely to spend their time on social media, and less likely to spend time online in general

Somewhat less likely to consume any media, including television, radio, newspapers, and magazines

What are the consumption behaviors that Doubters are particularly into? They’re significantly more likely to read Soap Opera Digest, FamilyFun, Seventeen, TV Guide, InStyle, and…Playboy. (The issue of trust in centerfolds goes unexplored in this study.) They watch We TV, TLC, and “Married at First Sight.” They drive Chevys, have cell service through MetroPCS, and eat at Bob Evans.

At the same time, though, “Doubters were very much like the rest of the country demographically, in terms of age, gender, race, ethnicity, and religion.”

Debates about media trust often end up debating a list of things news organizations could do in order to regain it. (Greater transparency! Events in the community! Solutions journalism! Um, blockchain!) But this data suggests that, at least for a small but significant share of the American public, the mistrust goes far, far deeper than a better corrections policy or hiring an ombudsman. It’s a deep-seated, core belief.