IQ skepticism: debate and objections

Interesting exchange by @tracewoodgrains with @IanMalcolm84

While almost nobody truly believes in a “blank slate,” it pervades education policy, as with the disaster of No Child Left Behind that punished teachers aiming to help the most disadvantaged students and smart kids wishing for No Child Kept Behind alike.

Agree. The evidence that some people are perhaps born to be more successful than others impugns on the notion of fairness and the ability of social interventions to promote equality, which explains the enduring appeal of blank-slate approaches to education policy. If only the right levers and buttons can be pushed, equal outcomes can be generated.

IQ has and will continue to be a touchy subject. Discourse about IQ is like being stuck between a rock and a hard place. The red pill of ‘IQ realism’ is hard to swallow, but denial is also bad. Or having to choose between realistic but low expectations vs. setting people up to fail.

The main issue is, critics argue that IQ is contrived and that it does not reflect reality: “It’s just a number!” Yes, IQ may seem contrived or oversimplified, but so are efforts to downplay it too. IQ skeptics have to create unrealistic scenarios or rationalizations for why IQ is not that important even when these fly in the face of reality.

It reminds me of the critcism over the BMI: “You cannot measure fatness with a single number! Bodybuilders have obese BMIs!” Yes, but bodybuilders are a tiny percentage of the population. The vast majority of people with BMIs above 30 are in fact fat. All you’re doing is creating your own useless contrived counterexample to refute how you believe BMI to be contrived.

He gives this contrived example:

Education matters. There is no IQ test on which you would score the same when you were raised as an impoverished subsistence farmer as you do in your current environment, no direct way to access “true” aptitude wholly detached from learned skills, and no sound reason to take patchy world IQ estimates as hard indicators of genetic potential.

Except who does this apply to? We’re talking cognitive differences in which it can be assumed that the environment is mostly held constant, not comparing extreme poverty to extreme wealth (and even so, wealthy blacks still underperform poor whites on standardized tests). In a typical middle class American classroom setting, everyone is roughly the same age and has equal number of years of schooling, yet there is considerable individual variance of ability.

I am sure everyone can relate to there being the ‘smart kids’ who seemed to pick up concepts faster than others, that could only be attributable to something innate like IQ. If those smart kids were tracked to adulthood, they would probably be overrepresented in high-IQ occupations at a rate that cannot be dismissed as chance. In fact, someone did this. Most debates about IQ hit a dead end where the skeptic concedes or hints at the existence of this ‘thing’ that accounts for wildly differing individual abilities, that happens to be similar to IQ but is not actually IQ. Or IQ by another name.

What does male variability actually look like in practice? The 2008 paper “Global Sex Differences in Test Score Variability” gives some idea by examining how 15-year-olds score on the PISA, an international education assessment test. Males are dramatically overrepresented at the low end and underrepresented at the top end in reading, and dramatically overrepresented at the top end in math while having comparable representation to women at the bottom end.

IanMalcolm84 is not entirely wrong. At worst, he’s giving a half-truth, but the official Wiki, which despite Wikipedia’s biases, still leans in favor of males having more variability. The majority of supplied evidence, such as meta studies, points to males having more variability for a wide range of abilities.

“There is no true ‘culture-free’ IQ test.”

Digit span is as close as culture-fair as possible, and Blacks tend to do much worse at backward-recall compared to forward-recall versus Whites on the same test. Moreover, the black-white IQ gap is as wide on matrix tests, which are considered more culturally fair, as it is on other type of test, such as verbal. Because the burden of proof of creating a sufficiently culture-free IQ test cannot ever be met in practical terms, it follows that racial IQ differences cannot be real. And if such a test existed, the implication is there would be no differences. Which by that logic, why have IQ tests at all? The logic is circular and presupposes its conclusion.

Additionally, blacks are highly overrepresented in many cultural domains such as sports, acting, and commercials, yet still perform much worse than immigrants, who have much less cultural immersion. There are many stories, like on Reddit and Hacker News, of first and second generation immigrants landing good tech jobs. Given the high rejection rates, what possible environmental attributes do they have over the other 99% of rejected applicants, if not IQ?

No one cares that much if Blacks are overrepresented at sports, but it’s a huge deal when Whites and Asians are overrepresented in STEM subjects or tech jobs. So it’s not as if society is totally or always opposed to racial inequality or differences. Diversity only goes one way, in much the same way other countries can have closed borders but America must let everyone in.

From @Philosophi_Cat

Regarding low national IQs, unintelligent people can be quite smart given their circumstances. 80-90 IQ homeless people are ingenious at finding ways to procure drugs or alcohol without having to work. But they are not smart enough to create better circumstances. Through repetition, low-IQ people can be competent at simple, domain-specific tasks and thus appear smarter than predicted by their scores. They make great specialists provided the task is simple enough and they have enough time to hone their skill, but they cannot just make the leap to something like getting a GRE or learning how to do fractions.