With the sphericity of the Earth empirically established by the ancient Greeks more than 2,000 years ago, it is difficult to believe that there are still holdouts. Yet, as a reporter for Vice recently observed, “If a flat Earth conference in Edmonton, Alberta, of all places, can pull in over 200 people … I think we may be underestimating the size of the movement.”

How prevalent is flat-Earthery? In a previous column, we examined what seems to have been the first systematic attempt to assess the American population’s views on the shape of the Earth, a YouGov poll conducted in February 2018. According to YouGov’s report, when asked, “Do you believe that the world is round or flat,” 2 percent of the 8,215 respondents chose “I have always believed the world is flat.”

When we asked YouGov for the data, however, we received a spreadsheet reflecting data for 10,374 respondents, of whom only 1.28 percent preferred the always-a-flat-earther response. Unfortunately, YouGov was unable or unwilling to resolve the discrepancy, making it impossible for us to reach a firm conclusion about the actual size of the flat-Earth movement on the basis of the poll.

Whether 2 percent or 1.28 percent, it’s still disturbingly high. So it’s not surprising that readers of our previous column were skeptical of the poll’s results—suspicious about the way in which the participants were selected, critical of the question used to assess belief about the shape of the Earth, and speculating about the possibility of insincerity skewing the outcome. So let’s take a look.

First, because YouGov’s poll was conducted online, a number of commentators expressed concern that a poll open to all comers would be vulnerable to a selection effect: flat-Earthers would be more likely to notice and answer the poll and to recruit allies to respond to it. That is a reasonable concern, but not applicable. YouGov’s polls are only administered to members of a prescreened panel of respondents.

Second, a lot of the skepticism centered, naturally, on the wording of the possible responses that YouGov offered to the respondents:

  • I have always believed the world is round.
  • I always thought the world is round, but more recently I am skeptical/have doubts.
  • I always thought the world is flat, but more recently I am skeptical/have doubts.
  • I have always believed the world is flat.

Responses that did not fall into those categories (7 percent in the report; 5.19 percent in the spreadsheet) were recorded in the Other/Not Sure category.

It is a serious flaw that the offered responses combine two questions: what respondents have thought in the past about the shape of the Earth (about which they may be unsure or unreliable) and whether they have recently entertained doubts.  So it isn’t entirely clear what the responses indicate about the current beliefs of the respondents with doubts.

It is not a serious flaw, however, that the offered responses assume the only relevant views are that the Earth is round and that the Earth is flat. Other views (such as the view that Earth is cubical, à la Scotland L. Moore’s planet of Aoicicinori) are probably not prevalent enough in the American population to be worth recording in their own category distinct from the Other/Not Sure responses.

It is a flaw, but probably not a serious flaw, that “round” and “flat” are imprecise. Technically, the Earth is not round, but rather a bumpy oblate spheroid. But it is unclear to what extent YouGov’s respondents were deterred from expressing acceptance of a round Earth by such considerations; it is likely that they generally discerned the pollster’s intention and selected the response closest to their view.

Third, in light of reports that younger people were less likely to prefer the always-a-round-Earther response, it was widely speculated that their responses were frivolous or ironic. Perhaps, but no evidence for it is apparent. Moreover, the most flagrantly insincere response would have been always-a-flat-Earther, while younger people were instead more likely to offer a response categorized as Other/Not Sure.

Perhaps it’s easier to admit the existence of flat-Earthers in light of a 2016 survey finding that 27 percent of Americans don’t accept heliocentrism, 48 percent don’t accept common ancestry of humans and non-human animals, and 61 percent don’t accept the big bang. Clearly, whether due to ignorance or ideology, the scientific consensus is not always accepted—so why not about the shape of the Earth?

If it’s difficult to believe that people embrace a flat Earth, it’s going to be difficult to trust a poll that claims to validate that belief. There were genuine flaws in YouGov’s poll, and there is clearly room for improvement. Still, along with the evidence from such events as Edmonton’s, the poll credibly indicates that—to parody the famous editorial about Santa Claus—yes, Alberta, there are flat-Earthers.