Academic (or other) backgrounds of AH denizens

[Kumquatqueen]

IQ is utter bunk.

Em

On the contrary, IQ tests are remarkably consistent at measuring whatever it is that IQ tests measure.

What that is, how the fuck it relates to any useful type of intelligence, and why it keeps rising with every generation, are what we have no idea of and why IQ shouldn't be taken seriously.

Also mentioning IQ, without which test was used when, is completely meaningless. They've been re-normed every few years.

Everyone I've ever met who has mentioned their IQ has been a total wanker. Denizens of this thread will be given benefit of the doubt until proven otherwise.

 

[EmilyMiller]

On the contrary, IQ tests are remarkably consistent at measuring whatever it is that IQ tests measure.

What that is, how the fuck it relates to any useful type of intelligence, and why it keeps rising with every generation, are what we have no idea of and why IQ shouldn't be taken seriously.

Also mentioning IQ, without which test was used when, is completely meaningless. They've been re-normed every few years.

Everyone I've ever met who has mentioned their IQ has been a total wanker. Denizens of this thread will be given benefit of the doubt until proven otherwise.

Mine was the TL-DNR version .

Em

 

[Tayloroc]

IQ is utter bunk.

Em

I agree.

Mine was the TL-DNR version .

Em

What I read is Lots of cultural biases. Dependent on motivation

 

[EmilyMiller]

On the contrary, IQ tests are remarkably consistent at measuring whatever it is that IQ tests measure.

What that is, how the fuck it relates to any useful type of intelligence, and why it keeps rising with every generation, are what we have no idea of and why IQ shouldn't be taken seriously.

Also mentioning IQ, without which test was used when, is completely meaningless. They've been re-normed every few years.

Everyone I've ever met who has mentioned their IQ has been a total wanker. Denizens of this thread will be given benefit of the doubt until proven otherwise.

Actually worse than quoting your own IQ as some badge of honor is saying historical figures had high IQs.

IQ proponent: Motzart obviously had a high IQ

EM: But the tests were only developed in the 1960s (I think, recently anyway, can’t be bothered to check)

IQ proponent: But he was a genius.

EM: And…

IQ proponent: So he must have had a high IQ

EM: [exit stage left, shaking head sadly]

Em

 

[Tayloroc]

Correlation….

 

[EmilyMiller]

What I read is Lots of cultural biases. Dependent on motivation

Care to elaborate?

Em

 

[EmilyMiller]

Actually worse than quoting your own IQ as some badge of honor is saying historical figures had high IQs.

IQ proponent: Motzart obviously had a high IQ

EM: But the tests were only developed in the 1960s (I think, recently anyway, can’t be bothered to check)

IQ proponent: But he was a genius.

EM: And…

IQ proponent: So he must have had a high IQ

EM: [exit stage left, shaking head sadly]

Em

The other circular argument is of course:

IQ proponent: IQ is correlated to intelligence

EM: So how do you quantify intelligence?

IQ proponent: Using IQ

EM:

 

[Tayloroc]

Testing is dependent on how much the person taking the test cares about the results.

 

[EmilyMiller]

Testing is dependent on how much the person taking the test cares about the results.

Oh I see.

Em

 

[Tayloroc]

Oh I see.

Em

But not my area. Just what I see in my classroom

 

[SimonDoom]

IQ is utter bunk.

Em

That's not what the evidence shows. It has to be put in perspective. It's a proxy. It correlates positively and significantly with a wide variety of aptitudes/skills/stuff that we generally consider "intelligence." It's flawed, and the correlation isn't exact, but that doesn't mean it's bunk.

If you were to take 100 people who tested 100, which is usually considered the mean IQ, and 100 people who tested 130, which is usually 2 standard deviations above the mean, you would see a very wide divergence in the two sets of people based on almost every possible measure of what you might think of as intelligence. Within each set there would be a very high degree of variability, but the mean differences would be extremely significant. We may not completely understand exactly how nature/nurture contributes to IQ, but that doesn't mean it's bunk.

 

[AwkwardlySet]

IQ is very much a valid thing, and it has been confirmed by decades of data. The way it is measured though, could definitely be improved. Some people use it as a trophy, (especially popular among actors and public figures) and some of them have been taking classes to "Learn how to do better on an IQ test" in order to reach a certain threshold and brag about their IQ, or to be able to join Mensa for example.
Modern society is shifting towards removing anything that might be considered as a discrimination, even in cases where it doesn't make any sense - like in the case of IQ, and that is why some other mostly pseudo-scientific theories have appeared (like Gardner's theory of multiple intelligences), so we can all be equally smart, more or less. What's funny is that, if we are all smart, then nobody is smart.

 

[EmilyMiller]

t correlates positively and significantly with a wide variety of aptitudes/skills/stuff that we generally consider "intelligence."

Can you tell me what the robust metrics are for these aptitudes/skills/stuff we generally consider “intelligence”?

You need two metrics to establish correlation, else it’s just “stuff we think”.

To cut to the chase, IQ has been twisted into a measure of intelligence (it was meant to be a way to identify children in need of remedial schooling). But there is no agreement on what intelligence is and no way to measure it directly. If you can neither define nor measure something, you can’t establish correlation.

Example. IQ is correlated to SAT scores. But d’oh! SATs are basically long-form IQ tests. Of course they are correlated and of course both suffer from the same biases with respect to - for example - ethnicity.

What has been established is that IQ (when assessed properly by professionals, not on-line tests) is self-consistent. But we don’t know that it measures anything. A number of people have a hunch that it does, but hunches don’t count for much,

Em

 

[EmilyMiller]

IQ is very much a valid thing, and it has been confirmed by decades of data.

Can you point me to one data set that confirms this?

Em

 

[XerXesXu]

IQ is utter bunk.

Em

IQ tests are a gatekeepers' test - 'Are these people like me?' Academics (before WW1) invented them to identify bright people - like themselves. As KQQ points out in #77, they undergo a great deal of re-norming to remove the most recently revealed biases. Many people show deep, specific intelligence. Many business people, scrap metal merchants, second-hand car dealers, criminals and even corner shop owners are very good at making money, and that specific intelligence tends to generalise to other matters, though it's not reflected in IQ scores. Often, these people have been poorly educated and poorly motivated to learn. Their abilities correlate highly with 'instrumentality' - what you need to know to achieve your specific ambitions.

It pays to be good at intelligence tests. The 11+exam was an intelligence test I passed with an A. That meant the local authority would pay for me to go to a selective school. I passed the school entrance exam, which was a conventional exam, and proceeded through the A stream to Sixth Form then on to university. I was the first person in my family (all those in my year were) to go to university, and then be the first to do lots of other things.

 

[AwkwardlySet]

Can you point me to one data set that confirms this?

Em

I wish I could, but it would require some tedious searching. I remember reading more than one study about it years ago. I've read those studies, triggered by the morons in my educational system who were for some reason personally fascinated by Gardner's theory, and since I recognized the ridiculousness of his theory before I even delved deeper into the whole thing, I felt the need to read more. It was quite some time ago so yeah...

 

[EmilyMiller]

IQ tests are a gatekeepers' test - 'Are these people like me?' Academics (before WW1) invented them to identify bright people - like themselves. As KQQ points out in #27, they undergo a great deal of re-norming to remove the most recently revealed biases. Many people show deep, specific intelligence. Many business people, scrap metal merchants, second-hand car dealers, criminals and even corner shop owners are very good at making money, and that specific intelligence tends to generalise to other matters, though it's not reflected in IQ scores. Often, these people have been poorly educated and poorly motivated to learn. Their abilities correlate highly with 'instrumentality' - what you need to know to achieve your specific ambitions.

It pays to be good at intelligence tests. The 11+exam was an intelligence test I passed with an A. That meant the local authority would pay for me to go to a selective school. I passed the school entrance exam, which was a conventional exam, and proceeded through the A stream to Sixth Form then on to university. I was the first person in my family (all those in my year were) to go to university, and then be the first to do lots of other things.

I agree that IQ correlates with an ability to take exams. Because they are basically the same thing. I assume the 11+ is a UK thing, forgive my US-centricity. I also assume that the people who do well at it are (tautologically) those who are good at sitting a certain type of test. It’s all circular, which I think you were also saying.

Em

 

[EmilyMiller]

I wish I could, but it would require some tedious searching. I remember reading more than one study about it years ago. I've read those studies, triggered by the morons in my educational system who were for some reason personally fascinated by Gardner's theory, and since I recognized the ridiculousness of his theory before I even delved deeper into the whole thing, I felt the need to read more. It was quite some time ago so yeah...

I’m not trying to be an asshole. Trust me I can be a lot more of an asshole if I try. But my point is that IQ is correlated to all sorts of things, but no correlation has been established between those and intelligence, which is a much more slippery concept.

So - at best - we have some self-consistent proxies, but no robust link between any of them and intelligence .

Em

 

[AwkwardlySet]

I’m not trying to be an asshole. Trust me I can be a lot more of an asshole if I try. But my point is that IQ is correlated to all sorts of things, but no correlation has been established between those and intelligence, which is a much more slippery concept.

Em

I didn't take it as such I can see this is an infuriating topic for you. Still, IQ and Intelligence are abstract terms we use to explain the, sometimes overwhelmingly huge, difference between people's abilities to act in new situations, or to "do well in school". These terms are flawed no doubt, and one probably doesn't translate directly into the other, but it seems that data from decades of IQ and psychometric tests point out that there is one generalized IQ that governs the abilities which we traditionally tie to intelligence. It is possible that there is some fundamental fault in the whole process and that some future data will completely contradict everything we got so far, but as with any other scientific theory, we have to work with what we have and make some deductions and conclusions, because the amount of data seems to be quite impressive and this theory has been around for a long time, so there was enough time for someone to make a serious dent in it, yet no one really did.

 

[SimonDoom]

Can you tell me what the robust metrics are for these aptitudes/skills/stuff we generally consider “intelligence”?

You need two metrics to establish correlation, else it’s just “stuff we think”.

To cut to the chase, IQ has been twisted into a measure of intelligence (it was meant to be a way to identify children in need of remedial schooling). But there is no agreement on what intelligence is and no way to measure it directly. If you can neither define nor measure something, you can’t establish correlation.

Example. IQ is correlated to SAT scores. But d’oh! SATs are basically long-form IQ tests. Of course they are correlated and of course both suffer from the same biases with respect to - for example - ethnicity.

What has been established is that IQ (when assessed properly by professionals, not on-line tests) is self-consistent. But we don’t know that it measures anything. A number of people have a hunch that it does, but hunches don’t count for much,

Em

This isn't true. There's an abundance of evidence that IQ correlates positively with WHATEVER you regard as intelligence. Memory. Analytical ability. Processing speed and ability. Computational ability. Knowledge. Financial success. Whatever.

If you want to take a radical skeptic's view of intelligence--that it isn't anything and it doesn't exist--OK. I think that's a foolish view. To me, based on everything I've experienced and read, intelligence obviously IS a thing. That it takes many forms and is difficult to define does not mean it doesn't exist and that it is not useful to talk of its existence or try to measure it.

It needs to be kept in mind that tests are a tool. They're not a measure of "intrinsic worth," whatever that is (probably nothing). They're a useful but imperfect tool for sorting people. Sorting people is a useful thing to do, and that's why we do it. Higher education institutions do it because it's useful to try to get the "smartest" students at their institutions. The military uses tests to sort people by the task they should do. These tools are rough and imperfect, but the evidence is pretty strong that they are not bunk and that they are useful.

I'm not a psychologist or an expert on this subject, I admit. I don't think it's necessary to demonstrate one's expert bona fides in a casual social media forum, which is what this is. So I'm not going to go to a lot of trouble to dig up all the stats and cites that support this. But I think you're wrong, and I think you're wrong from my everyday experience as well. In my experience, and based on what I've read, there is a significant though imperfect correlation between test performance and whatever we think intelligence is.

 

[EmilyMiller]

I didn't take it as such I can see this is an infuriating topic for you. Still, IQ and Intelligence are abstract terms we use to explain the, sometimes overwhelmingly huge, difference between people's abilities to act in new situations, or to "do well in school". These terms are flawed no doubt, and one probably doesn't translate directly into the other, but it seems that data from decades of IQ and psychometric tests point out that there is one generalized IQ that governs the abilities which we traditionally tie to intelligence. It is possible that there is some fundamental fault in the whole process and that some future data will completely contradict everything we got so far, but as with any other scientific theory, we have to work with what we have and make some deductions and conclusions, because the amount of data seems to be quite impressive and this theory has been around for a long time, so there was enough time for someone to make a serious dent in it, yet no one really did.

Things I accept: as valid:

1. IQ is correlated to achievement in exams
2. IQ is correlated to level of income
3. IQ is correlated to social status

Things I don’t accept:

1. IQ is correlated to intelligence
2. That we have any of a generally accepted definition of intelligence, a full understanding of intelligence as a concept, a way to measure intelligence directly

I wouldn’t use your definition of intelligence, which I think is circular. I’d say intelligence is the ability to figure out stuff. The problem is that the stuff could be such a wide range of things as to become almost meaningless.

Take Musk. Is he a savant, an idiot savant, or just an idiot. He’s all three. There are different types of intelligence and we currently have no way of boiling this down to a number. You need a number to establish correlation.

TL-DNR these is no SI unit of intelligence, that’s because we can’t measure it directly.

Em

 

[EmilyMiller]

This isn't true. There's an abundance of evidence that IQ correlates positively with WHATEVER you regard as intelligence. Memory. Analytical ability. Processing speed and ability. Computational ability. Knowledge. Financial success. Whatever.

My simple point is that none of those things is intelligence. They are all things that some people might associate with intelligence, but others might not. The Nobel Laureates I have met didn’t care much about financial success for example.

We have no agreed definition of intelligence, no direct measurement of it and so cannot possibly have anything whatsoever correlated to it.

Em

 

[TxRad]

Then there is the matter of all the narcisists and murders who are members of Mensa. It does not compute or does it.

 

[EmilyMiller]

Final point in this and then I’m off to either have sex or write about sex…

Hypothesis: IQ (A) is positively correlated to Intelligence (B)

Demonstrate for the examiner:

1. A reliable metric for A (shouldn’t be hard)
2. A reliable metric for B (advanced placement)
3. A sufficiently large population of observations {(a,b)} in A x B (honors class)
4. A correlation between the two metrics (Nobel Prize)

 

[Tayloroc]

Then there is the matter of all the narcisists and murders who are members of Mensa. It does not compute or does it.

I think it does. Being “smart” doesn’t make you moral