Hacker News Reader

Comment by contravariant | Hacker News Reader

Anyone with some domain knowledge willing to pitch in whether it's correct people are equating IQ with education in this thread?

View on HN

From what I've read is that IQ is largely genetically determined, With all the complexities that genes bring with them. And its very hard to increase intelligence in a meaningfull way. But whats important is that its very easy to make someone really really dumb. Like the disaster that was leaded gasoline causing an estimated loss of 2 to 8 iq points on a test.

Education absolutely increases IQ test scores but wether its an increase in intelligence itself whatever that may be is questionable.

An IQ test cannot distinguish between a subject who originally came up with a problem-solving technique due to innate intelligence, and one who has already seen that type of problem and learned the problem solving trick from someone else (possibly from having challenged numerous IQ tests).

There is some alignment between IQ tests and academic work; an academic background gives you a few tricks for solving those kinds of problems.

It all depends on how define intelligence, If by intelligence you mean the ability to reason about and solve complex problems then education definitely improves intelligence. However, if you choose to define it as something innate then by definition education does not change that. I think the former is a far more useful definition, its more important what a person can actually do as opposed what they could do in some hypothetical world

The problem I see is that intelligence is at large innate, someone with a 110 IQ bracket can't realisticly retest at an IQ of 140 unless you're training purely on IQ tests. This puts a severe upper bound on what anyone can 'solve trough education' as someone with a first tested IQ of 140 requires zero education to achieve a retest of 140 AND has the ability train or educate themselves even further. The IQ test itself is as real of a problem someone can solve as any other hypothetical problem. So if you choose to define intelligence by education, then you're still selecting for innate intelligence plus someones work ethic. Which to me is silly because the education system itself already selects for that and IQ tests are carefully designed for innateness which is why retests or learning for an IQ test bars you from a valid score. I'm sure Intelligence is somewhere inbetween but we can barely solve for one variable at a time.

A guess is that quality education, like encouraging the curiosity to read and tinker, has a bootstrapping effect.

So the Flynn effect (rises in IQ to a certain point in time) has been demonstrated in many many many cultures with many tests, many of which are difficult to explain in terms of formal western education. Like nonverbal pattern recognition tests given to traditional tribes in jungle areas. It's one of the mysteries about the Flynn effect and why it's so difficult to explain.

It's also not just the case that the "bottom is being raised", but also that the whole distribution was shifting.

I think these kinds of things make it unlikely education is explaining at least the rise part of the curve. To the extent the fall is in a subset of these types of measures of settings it also might not be (but it could be).

The more recent decline has been less well-documented although there are multiple studies using multiple measures that have shown this, so I believe it is a trend at least in some regions of the world (developed, western). I have a fuzzy memory that the decline is stronger with verbal measures (as opposed to nonverbal) but I might be misremembering that.

Education is strongly related to IQ/g/general cognitive ability but they're not perfectly correlated. I think when I looked this up a week ago or so, the best estimates were like 0.50 in a general population sample.

Clinically a discrepancy between IQ and educational achievement is important, as it points to someone not getting resources they need etc.

So yes, they're related, and yes, education could explain general population trends over time, but no they're not the same.

I just read this passage from the short story "Flowers For Algernon":

"I'm not sure what an I.Q. is. Dr. Nemur said it was something that measured how intelligent you were--like a scale in the drugstore weighs pounds. But Dr. Strauss had a big argument with him and said an I.Q. didn't weigh intelligence at all. He said an I.Q. showed how much intelligence you could get, like the numbers on the outside of a measuring cup. You still had to fill the cup up with stuff."

Maybe that stuff is education.

Here[1] there's a summary of the evidence for and against IQ.

[1]https://www.reddit.com/r/slatestarcodex/comments/oh6115/perf...

All that verbiage and the best he can do with respect to establishing causality is an offhand remark that causality is “heavily implie(d)”. Controlling for confounders won’t necessarily recover estimates of causal impact.

At the least I’d like to see something like an instrumental variables analysis.

Reply on news.ycombinator.com