Are Joe Biden and Taylor Swift working in cahoots? The late-night host Seth Meyers posed the question directly to the president Monday night, asking him to “confirm or deny that there is an active conspiracy” between him and the pop superstar.

“Where are you getting this information?” Biden responded. “It’s classified.”

The two were joking about the latest conspiracy theory that’s been bubbling under the surface for the last few weeks — that Swift, her tremendous popularity, and her saturated media coverage in the lead-up to the Super Bowl (she’s dating Kansas City Chiefs tight end Travis Kelce) is somehow a government psyop to influence American minds into supporting Biden’s reelection effort.

The theory has bounced around the conservative media echo chamber and even garnered some mainstream coverage. After being promoted on Fox News by host Jesse Watters, it spawned a field day of coverage on network news, daytime talk shows, and national radio.

It’s unclear what the origin of this conspiracy theory is, but one poll helped boost its reach. In the days leading up to the Super Bowl, Monmouth University asked whether respondents had heard about the idea of Swift being involved in a “covert government effort to help Joe Biden win the presidential election” and “Do you think that a covert government effort for Taylor Swift to help Joe Biden win the presidential election actually exists, or not?”

Thus did a viral conspiracy-theory joke — it’s still unclear how serious it ever was — go fully mainstream, all the way to a late-night TV show and the president himself.

The treacherous slope of a Taylor Swift poll

Here’s the thing about polls: Many times, they don’t really tell us what we think they’re telling us. Sometimes we have to peer deeper to find out what they’re actually saying.

The actual Swift conspiracy theory isn’t one unified concept. As my colleagues at Vox have explained before, it’s actually a whole bunch of ideas about Swift and her popularity being somehow artificially engineered and a sign of some covert effort to influence minds.

The idea of this being political was made more popular in elite right-wing spaces — pushed by failed GOP presidential candidates on social media and Jesse Watters in primetime, all to suggest that Swift is part of a psychological operation being used by the Pentagon or the federal government to convince her followers to support Democrats like Biden.

For many people, the Monmouth poll’s findings were jarring. Some 18 percent of respondents said that yes, they did believe that a “covert government effort for Taylor Swift to help Joe Biden win the presidential election” existed.

That number mostly consists of who you might expect: 71 percent of believers identify as Republicans, and an even greater percentage, 83 percent, say they’ll likely back Trump this fall. And the numbers for those who had heard of this kind of conspiracy theory at all were similarly eye-opening: 46 percent of Americans had been exposed to the idea.

No wonder the results went viral.

But dig a little deeper and the poll results can start to make you question your priors. A decent chunk of those who said they believed in the theory were actually unaware of it before Monmouth contacted them, leading to a fundamental question: How many believers actually “believe” in such a conspiracy theory?

With any poll of wacky ideas, it’s important to keep in mind the concept of the “Lizardman’s Constant” — the idea that a certain number of people being polled on a weird topic will probably not answer sincerely — and that asking questions on more ridiculous topics will probably get you more ridiculous answers.

There are two poll results that should cause some introspection. First, among those who believe Taylor Swift is a pro-Biden psyop, 42 percent had not heard of the conspiracy theory before Monmouth contacted them. And of those who were previously aware of it, it turns out more were Democrats (56 percent of them) than Republicans (46 percent).

Though the Swift psyop conspiracy theory may have originated among conservatives, it appears to have spread more widely through liberal and Democratic social networks and mainstream media — like through mainstream coverage of right-wing media, and eventually coverage of the poll and Meyer’s Biden interview. According to Patrick Murray, Monmouth’s polling director, that’s a significant detail, since it’s showing how viral ideas spread.

“Democrats are more likely to be aware of it than Republicans, I think in part because the idea [that Republicans believe this] has gotten more currency on the left,” Murray told me. So it’s not that Democrats believe the theory, “but they’re hearing more about this, and in turn they’re actually talking about it more.”

At the same time, the poll indicates that although comparatively fewer Republicans have heard about this conspiracy theory, they are much more likely to believe it to be true — about a third of respondents who believe the theory are Republicans. That doesn’t mean that they’re all in agreement about the specifics of this supposed deep-state operation. But it does suggest a kind of rally-around-the-flag effect for Republicans — who may be more willing to “incorporate this in some way, shape, or form into their belief system” about American politics, pop culture, and media, according to Murray.

That idea also frames that 42 percent of people who said the conspiracy exists but also said they hadn’t even heard of the idea before pollsters contacted them. They could be the kind of people who genuinely believe in a conspiracy. They could also be the kind of people who are communicating an emotion — not that Taylor Swift is literally an agent of the deep state, but that she is a stand-in for a worldview suspicious of American politics, loyal to Trump, and believe the system is “rigged.”

Murray said that this cohort skews Republican and that some respondents might not have known what they were agreeing to, but responded affirmatively because that’s what fit with their partisan loyalty.

“There’s always going to be an element of ‘To what extent are you agreeing with something because you wholeheartedly believe every aspect of it or because it helps further an agenda that you have?’” he said.

There’s a comparison to be drawn here with how people respond to polls about the economy, Murray suggested. “How you feel about your own economic situation has a lot more to do with politics now than it ever has. The lens through which people view their own economic situation has a lot more to do with their political identity than it does with their actual financial stability.”

I reckon a high percentage of respondents would still say yes if you substituted Taylor Swift's name with practically any other celebrity https://t.co/BC8QAd3du5

— G Elliott Morris (@gelliottmorris) February 19, 2024

Wacky questions bring out wacky respondents

And finally, there’s the Lizardman’s Constant: Some number of these Swift psyop believers might be messing with the poll.

Lakshya Jain, an analyst at the election modeling website Split Ticket, is one of the election watchers who was skeptical of the poll. He said the survey reminded him of the discourse around Public Policy Polling’s conspiracy theory research in 2013 that found a not insignificant number of Americans believing a slate of wacky ideas, like whether shape-shifting lizard people exist, whether Barack Obama was the Antichrist, and whether the moon landing was faked.

That poll spawned the concept of the Lizardman’s Constant on Scott Alexander’s blog Slate Star Codex — the idea that in any given poll, a percentage of responses are not actually genuine. According to Alexander, Lizardman’s Constant tends to be 4 percent — the number you should subtract a wacky result by to get closer to the truth.

“What we see is that if you ask any weird question, you’ll get a substantial portion of people agreeing to this kind of weird response,” Jain told me. “But if you ask obviously ridiculous questions in a survey, you’re going to get some people who agree.”

Pollsters do a lot of work to correct for that possible error, but it’s a helpful idea to keep in mind when looking at Taylor Swift conspiracy poll results. The actual number of Swift Psyop Believers might not actually be 18 percent of Americans.

But whether the number of true believers is 18 percent or 14 percent or 10 percent, the idea is out there — enough that even the president is leaning into it, and maybe feeding the conspiracy even more.

$5/month

Yes, I'll give $5/month

Yes, I'll give $5/month

We accept credit card, Apple Pay, and Google Pay. You can also contribute via