This is a bit of a niche topic, but I really hate scammers and eugenicists. Paul Cooijmans, a self-appointed expert in human intelligence from the Netherlands, has built a career of designing “high-range” IQ tests.

He insists that “childhood age-peer scores” are invalid, as are classic ratio IQ tests (for example, a score of 100 means that someone’s “mental age” is the same as their chronological age), and only adult standard deviation tests (using an SD of 15) can be used to rank people.

But this is a scam designed to get people to take his tests. (You can apply these arguments to any of the other absurd high-IQ societies on the internet that sell their own bespoke tests.) Here’s why:

  • IQ tests were originally designed for children. Not because individual differences in adult intelligence are irrelevant (and the relative skill level is usually constant), but these tests are designed to identify kids’ skills in school. That’s what Alfred Binet intended when he created these tests. They were made to help kids, not give people bragging rights. Adult tests are especially useless for identifying one’s learning potential, since they are typically used for clinical purposes, such as identifying whether someone has difficulty with executive functioning or short-term memory. Once you’re grown up, it’s much clearer what you can do. David Wechsler was clear that his scales were clinical tests.
  • Also, modern-day tests like the WAIS do not include high-range scores like this on their adult tests. Remember that he considers adult scores with a standard deviation of 15 as the only acceptable ones. But these tests are typically cut off at 160 and are geared towards the middle of the bell curve (roughly 70 to 130).
  • He calls the Stanford-Binet test a “childhood test.” The Stanford-Binet is an all-ages test. Older versions, such as Form L-M, are also known to be better at identifying both highly intelligent children and those with intellectual disabilities. (Full disclosure: I’ve taken both Stanford-Binet and Wechsler tests.)
  • As far as I know, all modern tests use age bands to rank people. If a 20-year-old takes the Wechsler Adult Intelligence Scale (WAIS), they will be compared with other 20-year-olds, not 18-year-olds or 40-year-olds. The same goes for the modern Stanford-Binet test and the Kaufman tests, and just about any other test you can find on the market. These are age-peer scores, just for adults rather than schoolchildren.
  • Cooijmans’s devaluing of childhood tests may be an artefact of his sexist views. People like Cooijmans believe that girls can be precocious (rather than intelligent or gifted) and plateau or regress to the mean in mid-childhood or adolescence, but boys take longer to mature, so their scores are valid at a later age. This is bullshit. The kinds of girls who show clear signs of high intelligence at a young age (existential concerns at 7, reading at 18 months, going to university at 10) are unlikely to drop to a 100 or 120 level as they get older.
  • Intellectual disability must be diagnosed before the age of 18. An adult test will not suffice for diagnosis. The same should apply to the right tail of the bell curve. Extremes in intelligence are a developmental phenomenon.
  • He is specifically attacking the older Stanford-Binet tests (specifically form L-M), which were the only mainstream psychometric tests to produce scores above 160, using a special formula that combines a ratio and a deviation score. The standard deviation of the L-M  is 16, rather than the 15 used on modern low-ceilinged tests like the dreaded Wechsler scales. If Stanford-Binet results, typically taken in childhood, are invalid predictors of someone’s intelligence, then they’ll fork over for Cooijmans’s homemade IQ tests. His disparagement of these tests is in his financial best interests. When people make claims like this and sell a competing product, follow the money. You’ll rarely be led astray.
  • Cooijmans’s tests can’t be used to join mainstream high-IQ societies like Mensa. You can’t take them to a neuropsychologist to show your intellectual functioning. You can’t do anything with these tests but brag about them online. They’re an intricate puzzle, not a real IQ test.
  • The WAIS and similar tests will probably bore the hell out of someone who has strong logical and pattern-recognition skills, since the subtests tend to be rather basic and a bit dull. Also, they include a lot of subtests that involve motor, visual and other non-intellectual skills. They are inaccessible for a lot of people.

And here’s where Cooijmans comes in. He says that only tests like the WAIS produce valid scores. These scores cut off most people who are drastically different from the norm. People who desperately want to prove themselves then spend money on Cooijmans’s tests. His business preys on people who want to prove their intellectual bona fides. This is nothing but a scam that preys on the relatively intelligent (but gullible), rather than the average person’s multilevel marketing or anti-cancer bracelets.