Ask A Genius 1259: Nicholas Taleb on IQ
Author(s): Rick Rosner and Scott Douglas Jacobsen
Publication (Outlet/Website): Ask A Genius
Publication Date (yyyy/mm/dd): 2025/02/14
Rick Rosner: Dave Chappelle gave the longest monologue in the history of SNL tonight and received a standing ovation. He mentioned regretting some of his past remarks—perhaps those made during the Trump era. I wish him luck and hope he gives the historically disenfranchised a fair chance. I also remember a snippet where he praised Jimmy Carter for boldly going into Palestine without the protection of Israeli forces. When was that? Early 2000, I believe—shortly after Chappelle left his show and began travelling the world.
On another note, Nicholas Taleb’s work is provocative; he flat-out calls Mensa a group of egotistical, blowhard losers and dismisses many IQ proponents, like Jordan Peterson and the authors of The Bell Curve (which, by the way, is over 40 years old now).
A proper interpretation isn’t just about interpretation—it goes beyond that. You have to read the paper because it doesn’t merely discuss interpretation; it dismantles the whole idea. It is far more thorough than simply criticizing the interpretation of IQ tests. It dismisses everything, including the very concept of IQ–an illegitimate measure. But you should look through the paper because it drives the point home from many angles.
The paper is fired up; the author is energetic.
Taleb makes the basic point that I had never fully appreciated: even a 0.5 correlation isn’t great. Since correlation is squared to determine the additional information it conveys, you need to square your correlation to see what extra insight you get.
It showed a U-shaped graph. It indicates that you obtain a noticeably helpful amount of additional information only in the super high correlations—0.8 and above—. He also points out that IQ hardly correlates with anything that isn’t itself. Although there is a high correlation between SAT scores and IQ scores, it isn’t as high as expected. Individuals retaking multiple IQ tests only achieve a self‑correlation of 0.8. That is much lower than the correlation you’d see with height or the time it takes to run 100 meters.
A self‑correlation of 0.8 means that there is only 64% information in any individual IQ score. When you use that loosely correlated number to compare it with other variables, it turns out to be a poor measure. He even goes so far as to say that the worst use of IQ might be in discussing the average IQs of entire countries.
In one study of the national average IQs of 185 countries, the paper shows that for 104 of those countries, there was no actual IQ data, and nobody was tested. Instead, the figures were extrapolated from racial demographics, which are circular, racist, and spurious. Thus, many of the things IQ is used for—especially by racist individuals—are utterly flawed.
He also discusses the asymmetry we mentioned yesterday. At the low end, IQ is correlated with poor performance. For example, if you have a test group where 20% of the subjects are deceased—scoring zero on an IQ test and zero on any performance measure—this creates false correlations. The asymmetry of IQ as a measure means that while a low IQ indicates impairment, there is no corresponding guarantee that a high IQ will ensure exceptional performance in areas not directly derived from IQ tests.
Being dead gives you a 0% score across the board. There’s no opposite extreme where being alive and smart guarantees a 100% score on every performance measure. Brain damage leads to poor performance in several areas, but the scale is not symmetric at the high end.
This lack of symmetry leads to many misconceptions, especially among people who do not know how to interpret statistics—which might include me since some of this information was new and somewhat surprising. Taleb has devoted much thought to statistics, probability, and those arcane situations that most statisticians overlook yet are very applicable to the real world. He wrote The Black Swan, which discusses rare occurrences that defy traditional statistical expectations; improbable events happen only rarely, so you must figure out how to adjust your models to account for these rare instances.
Much of what he brings up regarding IQ shows that many people in the field do not understand their discipline. Most statisticians do not grasp all the necessary subtleties of statistics. As for specialists in IQ—if there are any left making their living solely by working with IQ—that field has largely declined.
Interest in IQ has diminished, and perhaps mockery of it has increased. Although many people once earned their living during its heyday, that is no longer true.
Primarily, you’ll find this in the individuals who need to identify struggling kids and get them into a program. It’s done through mainstream, legitimately proctored tests that help people succeed. Given that most statisticians don’t understand statistics and its subtleties, the people working with IQ—such as those involved in getting gifted kids into specialized programs or the psychologists you mentioned who privately test children to qualify for gifted programs—charge around $500 to administer the test. (What’s the good one? Oh, Stanford‑Binet—is it? It’s notorious for its high scores.)
They charge $500 to write a report, but that might be only about 3% of their practice; most psychologists are meeting with people rather than administering IQ tests. There is no way these professionals fully understand the pitfalls of IQ and statistics. The field is rife with statistical ignorance and even self‐educated racists who have read The Bell Curve and listened to Jordan Peterson. These individuals are eight steps removed from any legitimate statistical knowledge. Telev has spent much time examining the field and realized its foundation is as flimsy as pudding. There’s nothing solid holding it up.
He makes two main points: first, IQ cannot do much of what its proponents claim it can, and second, the things it can do can be determined easily without resorting to IQ tests. There is, however, a potential weakness in one of his arguments. For example, if a child has learning difficulties, those issues might be more apparent when evaluated in high-level tasks rather than through IQ tests alone. Spending a few hours administering IQ‑type tests might sometimes help provide a clearer picture of a child’s deficiencies.
Nonetheless, he argues that IQ is mostly useless.
Last updated May 3, 2025. These terms govern all In Sight Publishing content—past, present, and future—and supersede any prior notices. In Sight Publishing by Scott Douglas Jacobsen is licensed under a Creative Commons BY‑NC‑ND 4.0; © In Sight Publishing by Scott Douglas Jacobsen 2012–Present. All trademarks, performances, databases & branding are owned by their rights holders; no use without permission. Unauthorized copying, modification, framing or public communication is prohibited. External links are not endorsed. Cookies & tracking require consent, and data processing complies with PIPEDA & GDPR; no data from children < 13 (COPPA). Content meets WCAG 2.1 AA under the Accessible Canada Act & is preserved in open archival formats with backups. Excerpts & links require full credit & hyperlink; limited quoting under fair-dealing & fair-use. All content is informational; no liability for errors or omissions: Feedback welcome, and verified errors corrected promptly. For permissions or DMCA notices, email: scott.jacobsen2025@gmail.com. Site use is governed by BC laws; content is “as‑is,” liability limited, users indemnify us; moral, performers’ & database sui generis rights reserved.
In-Sight Publishing 