How different are various IQ levels to each other, ie 100 vs 110,...?

In My Humble Opinion
101

Yes, practice effects are a big thing. That is one of the reasons that test preparation for the SAT (and similar) can be useful.

The test retest reliability means that your scores across multiple time points are all highly correlated, but not necessarily the same.

Because she is having trouble in school, my daughter took an IQ test in a clinical setting for exactly those reasons, Anybody talking to her would know she does not have an intellectual disability, but a standardized test can put a number on it, which communicates to anybody looking at her file that this child is having difficulties for reasons other than intelligence.

This is my opinion, and probably better suited for the “what do you believe without evidence thread,” but a lot of people hate the idea of IQ, because they hate individual differences, the idea that individual differences can be quantified, and that some of these differences can be important.

People are different. Some of those differences can be easily measured, such as height, weight, preferred way to pronounce GIF, and even IQ. Some of these measured differences can have tremendous consequences for the people who end up on the wrong end of the distribution.

That really upsets some people, who equate what I just said with racism and eugenics. To be completely fair, eugenicists love IQ and other measurable individual differences because it can give them an air of scientific legitimacy to support their racist views.

So, people will poke at IQ, g, and other things, pointing out legitimate flaws, exceptions, and just making stuff up in attempts to discredit it. We’ve all seen similar behavior around things like vaccinations, climate change, etc., and they’re often just as insightful as people asking “if we have global warming, why is it so cold outside?”

102

I suppose it depends on your definition of “smart”; but I’d say that, if you’re getting better at a particular kind of thinking, you are getting smarter, if only in one particular, limited way. (Kind of like, you’d be getting stronger if you were strengthening one specific muscle.)

103

No doubt. In this case I meant “smart” in the sense of the measurable, relatively static traits we’ve been discussing. I could’ve been more clear.

104

Some of the recent comments double on the possibility of a person with a measured IQ of 90 becoming a physician.

I will argue yes possible.

Slates are not blank and there are inherent differences in intelligences. (Yes the plural form is intentional.) Not sure how convinced I am of g per se but I am convinced that all tools designed to measure g are limited and minimally have significant cultural biases.

There are definitely people whose measured IQ on the accepted instruments are lower than their true intellectual capacities. And the other direction as well.

105

I think (and if I’m wrong, apologies for putting words in your mouth) that you’re suggesting that the IQ-83 guy might not really have a low IQ but is writing that letter in an attempt to discredit the concept of IQ. Certainly possible.

I think that IQ is real and important, but it’s not the only important thing, and of course something that is statistically valid is not always valid for an individual. The map is not the territory.

I’m sure that it’s challenging to have an IQ of 90, but I also think that the claims made in the 4chan screenshots are total nonsense.

Looking again at the link I shared, I hadn’t noticed before that it was written by Scott Alexander of Slate Star (and Astral Star) Codex. I respect his writing and thinking a good deal, and he’s a trained psychiatrist, so this stuff is at least a little in his wheelhouse.

It’s also worth spending more time with the chart at the end, which is from a legitimate scientific paper on IQ, and clearly is not consistent with the 4chan nonsense.

Kindergarten/elementary teachers have a 10th percentile IQ below 90. I am absolutely positive that ~15% of kindergarten/elementary teachers do not struggle to comprehend “how would you feel if you hadn’t eaten yesterday”.

Materials and design engineers have a 10th percentile just above 90. I’m not sure exactly which jobs count as material and design engineering, but I find it hard to imagine that ~10% of them can be done by people who struggle to contemplate a theory of mind for themselves a day ago.

106

Yes, thanks for stating it clearer than I did, that is what I mean. I don’t doubt he tested at an IQ of 83, if that’s what he claims. I once tested with an IQ of 1, and my brother with an IQ of 0. I thought the test was stupid, and only answered 1 question, my brother was told to only answer what he was sure of, so he didn’t answer any.

The point being, that a single score on a single test from a single subject doesn’t demonstrate much of anything. Like I said, it’s the equivalent of “how can there be global warming, it’s cold outside.”

Lots of scores on lots of subjects and finding lots of exceptions just proves that the correlation between IQ and whatever is not 1.0.

I think this quote from that article is very important for people to remember:

The chart perfectly demonstrates how IQ is both statistically reliable and individually unreliable. On average, intellectually demanding occupations like college professors have higher IQs than less demanding occupations like janitors. But individual janitors are sometimes higher-IQ than individual college professors. And almost every profession draws from a wide range of IQs. The average professor is pretty smart — but a nontrivial number have below-average IQs.

IQ, BMI, and other things can work well on a population level, but really fall down on an individual level. BMI can be a powerful predictor of heart disease in a study, but used for fat shaming on an individual level. IQ can be a strong predictor of success, and used as a basis to commit forced sterilization under evil eugenics plans.

So if I know nothing more than “that person’s IQ is 90” I’m going to say their a poor candidate for college professor. If all I know is “that person’s BMI is 32” I’m going to say they are at risk of heart disease.

It’s easy then to do the reverse, too. Look at a bunch of people in jobs that seem to require high IQs, and then find the ones that have records of low IQs, or find a bunch of people with low BMIs, but heart disease, and then write a gotcha article. None of that means IQ or BMI is invalid, or a useless predictor, just that the correlation is not 1.0.

Or another example. On average, men are taller than women. Not all men are taller than all women, so a short man and a tall woman doesn’t prove anything. It doesn’t mean the concept of height is invalid, or that “men on average are taller than women” is incorrect.

107

That’s very possible. It’s also possible that the test “writers” slapped something together and it spit out a very low number if the person only answered a few questions. If they didn’t anticipate people skipping questions (or thought that no answer meant the person couldn’t answer), they would assign 0 as though they missed it, etc.

I once knew a guy who took college courses to become a forest ranger. He was a nice guy but not the smartest person I ever knew. I wouldn’t be surprised if it took him six years to complete a four year degree, but a lot of people need some extra time. A lot of us thought he was crazy because the job doesn’t pay well, but it’s his life etc.

What if you took 100 guys who loved the park system, forests, etc., and who had bought into getting an education to realize their dream…and you also took 100 guys who loved medicine, health care, and who had bought into getting an education to realize their dream of becoming a doctor? Check back in ten years. How many became forest rangers, and how many became doctors?

I found this:

Many Americans may not know that the number of doctors is essentially set by Congress. Each year, the Centers for Medicare & Medicaid Services, or CMS, provides payments to hospitals to hire their residents—the crucial bridge that medical school students must take to becoming licensed physicians. Because the number of residencies is tantamount to the final number of doctors, the federal government has enormous leverage over the number of new physicians that enter medicine each year.

Source

And there’s that joke: if you and I are being stalked by a grizzly bear, I don’t have to outrun the bear—I only have to outrun you.

In other words, lots of folks could study to be forest rangers and all could pass. Doctors, no…there’s a limiting factor. Even if they’re all brilliant, that fixed humber of residencies apparently is your limit. And your fellow students are your competition. Got a 90 IQ, Mr. Hopeful Doc? Here’s your IQ 120 competition…he picks up things faster, has a lot of previous knowledge to connect it to, and he’s very quick thinking on his feet.

IANAD. Feel free to fight my ignorance.

108

If you’re interested in digging into the data and getting some insight into what confounding factors might be taken into account, here is a link to Robert Hauser’s original paper, Meritocracy, Cognitive Ability,and the Sources of Occupational Success. It is the source of the bar chart in the article. There are also a lot more charts of this type at the end of the paper.

109

Yes getting into medical school is a competition. There are tons of students starting college hoping to get one of those spots and only a very few of those students will end up as physicians.

The being chased by the grizzly bear analogy though is where your understanding is insufficient.

One, you are missing my point in the post you reference. Having higher g (assuming g actually exists as a thing other than as defined by your performance on the test stated to measure it) can certainly make it easier to get top grades and to test well on the MCATS. I am however not convinced that IQ test scores necessarily always measure whatever g is in an accurate manner. The test is not always a valid measure of whatever g is, and historically has been prone to cultural biases.

Then there is simply that higher intelligence alone is not the metric of who outruns the bear and who does not. I am 100% confident that there were many starting off with ambitions for med school the same college freshman year I made that choice with higher g than I have who did not make it into med school. And some with g less than I have who did. Yes there is some soft threshold of sorts on the low side, but above that there are many other differentiating factors and paths to being chosen for one of the spots.

Someone testing as 90 may still in reality be smart enough, and have other characteristics that serve them better in the competition than someone with slightly more g than they have.

A good doctor does need to be smart enough and above that there are many other factors that create a better doctor, and then those factors vary by specialty. Highest possible intelligence doesn’t necessarily correlate with being the best at the job. In fact one of the classic gags in med school is that you want to avoid both the physicians who were at the very top and the bottom of their class. Of course those of us who were neither were the ones telling that gag.

110

Pages 27 and following are specifically relevant to this discussion btw.

111

To nitpick myself, an I.Q. of 300 isn’t 20 standard deviations above the mean. It’s 13 and a third standard deviations above the mean. I’ll let someone else calculate how ridiculously improbable that is.

112

You’ve been there; I haven’t. And I know from reading many of your posts that you’re pretty circumspect in your thinking. Ignorance fought, thanks…I’ll have to ponder further.

113

Thank you. I tried to follow the link in the article but it was dead and I didn’t pursue further.

114

I was given an IQ test at 7 or 8 years old by the child psychologist who was treating me for depression, but I wasn’t told that’s what it was. It seemed like a series of puzzles and games, which wasn’t that different from some of the other therapy activities for little kids.