Over the years I’ve made many attempts to estimate the IQs of my readers. Let’s take a look down memory lane:

Gestalt test (Estimated mean self-reported IQ 127 )(U.S. norms)

Based on self-reported performance on a Gestalt test, I estimated my readers have a mean IQ of 123 with an SD of 16 or 19.5 (depending on how Gestalt IQ is calculated). But as a commenter noted, the Gestalt test was normed in Ontario which has a mean IQ of 106 (U.S. norms) and Canada as whole has an SD only 89% as large as America’s (Ontario’s SD is not known).

On a scale where the American mean is set at 100 with an SD of 15, Ontarians are likely 106 with an SD of say,13.4. Using the same U.S. norms, my readers would thus average IQ 127 (SD 14 to 17)

Professionally tested (Mean self-reported IQ 129)(U.S. norms)

Back in May 2017 I befriely ran a comprehensive poll of my readers which included the question What was your most recent score on an official professionally administered individual IQ test taken in the last 10 years? An astonishing 59% claimed they were professionally tested, and their self-reported scores had a mean of 129 with an SD of 21 (U.S. norms). Of course self-reported polls (even anonymous ones) can’t always be taken at face value. There is self-selection in who answers which questions. There are trolls self-reporting IQs as low as 40 or as high as 160+. In addition some of the tests might have been used outdated or non-American norms. Nonetheless, a beautiful bell-like curve emerged:

The SAT (Mean self-reported score equated to IQ 132)(U.S. norms)

Perhaps the single best data-set was from 2015 when I asked readers to select their SAT scores (verbal + math) along with whether they had taken the test before of after the April 1995 re-centering. When converted to IQ equivalents (U.S. norms) readers averaged 132 with an SD of 20. Since 91% of American readers who were old enough to have taken the SAT/ACT had actually done so, those who took the SAT seem representative of American readers as a whole and it’s unlikely that the distribution would change much if the 9% who had not taken it could be forced to do so.

Although the distribution does not form as bell-shaped a curve as the previous graph, there is little evidence of trolling. Till this day no one has ever claimed a perfect score on the pre-recentered SAT. Perhaps this is because the primary troll is “Mug of Pee” and he restrained himself from trolling this one poll out of reverence for the SAT which he considers the best IQ test.

Summary

Gestalt test: Mean IQ 127 SD 14 to 17

Professional IQ test: Mean 129 SD 21

SAT IQ equivalent: Mean 132 SD 20

Average of all three: Mean IQ 129 SD 19

Jonathan Wai method

Professor Jonathan Wai has argued that you can estimate how many top 1% minds are in a group by counting how many elite school alums are in that group. The elite school alums and the top 1% minds will not necessarily be the same people, but the frequency of one arguably predicts the frequency of the other. Assuming the group does not actively select for elite alums, this approach might work.

In 2015, a poll of my readers found that of the 61 respondents who were both American and old enough to be alums, an astonishing 62% claimed to be elite alums implying 62%of my readers have IQs of 135+ (U.S. norms).

I find that hard to believe, but it’s one more piece of evidence pointing to the mean IQ of my readers being somewhere around the +2 SD mark.