Where should the best and brightest go?

Quote function not working, @Astner:

There are lots of studies of this manner; they're very well replicated. The wikipedia article summarizes the evidence well.[url=https://www.ncbi.nlm.nih.gov/pubmed/23919982] This article elaborates on a related point, that the heritability of IQ increases with age, and that environmental effects matter a lot when you're young, but usually even out over the course of adulthood.

It's actually a really well replicated result in psychometrics that nonetheless is not well accepted by the general public for political reasons.

The other point of interest is that nobody has ever managed to permanently increase someone's general intelligence through training. You can cram for a test and then do better on it, but give them an unrelated test and they don't do any better than they would have before. Programs like Head Start produce improvements in standardized test scores, but only for a few years; they have long disappeared by middle school.

They can be raised in similar—but never identical—environments. They're not going to be in the same class in school, they're not socializing with the same people.

If one of the siblings has a friend that introduces him to a sport that he later develops into a career, then yes he's going to be different from the sibling who never had an interest for sports.

Regardless, we know your score on the SAT is better correlated with that of your biological parents than your adoptive ones.

Nature implies that it is genetic, which means that it's inheritable. But if brothers tend to turn out differently then it's not genetic.

You have it backwards. Something being heritable usually implies that it is genetic. Likewise, brothers turning out differently doesn't mean that it isn't genetic, since there is genetic variation between offspring.

--------------

Even if we step outside of the studies, I think it's incredibly unreasonable to think that natural aptitude doesn't play a role in people's skill set. It's difficult to square that with child prodigies, or with people who honestly try very hard at something, and receive good tutoring, but still can't seem to "get" it. Nobody's suggesting that the environment plays no role, but the opposite is just wishful thinking.

Originally posted by The Ellimist
This article elaborates on a related point, that the heritability of IQ increases with age, and that environmental effects matter a lot when you're young, but usually even out over the course of adulthood.

Originally posted by Astner
IQ is non-scientific because it's a test score that doesn't reflect what it's suggested to reflect. Moreover, the largest and most comprehensive MRI study ever conducted in the field of neurobiology suggests that the very idea that intelligence can be measured with a single number is wrong, [url=http://dx.doi.org/10.1016/j.neuron.2012.06.022]source.

Originally posted by The Ellimist
The other point of interest is that nobody has ever managed to permanently increase someone's general intelligence through training.

However, the volume of grey matter in the frontal lobe measured in MRI scans that are usually associated with general intelligence can be increased.

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3004979/.

Originally posted by The Ellimist
Regardless, we know your score on the SAT is better correlated with that of your biological parents than your adoptive ones.

Originally posted by The Ellimist
Even if we step outside of the studies, I think it's incredibly unreasonable to think that natural aptitude doesn't play a role in people's skill set. It's difficult to square that with child prodigies,

Originally posted by The Ellimist
or with people who honestly try very hard at something, and receive good tutoring, but still can't seem to "get" it.

Originally posted by The Ellimist
https://www.ncbi.nlm.nih.gov/pubmed/23919982 This article elaborates on a related point, that the heritability of IQ increases with age, and that environmental effects matter a lot when you're young, but usually even out over the course of adulthood.

Originally posted by Astner
IQ is non-scientific because it's a test score that doesn't reflect what it's suggested to reflect. Moreover, the largest and most comprehensive MRI study ever conducted in the field of neurobiology suggests that the very idea that intelligence can be measured with a single number is wrong, source.

Originally posted by The Ellimist
The other point of interest is that nobody has ever managed to permanently increase someone's general intelligence through training.

However, the volume of grey matter in the frontal lobe measured in MRI scans that are usually associated with general intelligence can be increased.

Source.

Originally posted by The Ellimist
Regardless, we know your score on the SAT is better correlated with that of your biological parents than your adoptive ones.

Originally posted by The Ellimist
Even if we step outside of the studies, I think it's incredibly unreasonable to think that natural aptitude doesn't play a role in people's skill set. It's difficult to square that with child prodigies,

Originally posted by The Ellimist
or with people who honestly try very hard at something, and receive good tutoring, but still can't seem to "get" it.

^ just to respond to your second post (EDIT: which has been deleted), general intelligence's validity comes from the positive correlation that can be observed between all cognitive tests. g can then be extracted via factor analysis. .i.e. there's a strong positive correlation between performance on verbal tests and math tests, math tests and physics tests, physics tests and job performance, and all of these things together.

The question of whether g is some underlying neurological structure or if it's just a statistical artifact is a little more debatable than the question of whether it's anything at all; it clearly comes from something, and it's clearly predictive of relevant real life outcomes.

Originally posted by The Ellimist
g can then be extracted via factor analysis. .i.e. there's a strong positive correlation between performance on verbal tests and math tests, math tests and physics tests, physics tests and job performance, and all of these things together.

Originally posted by The Ellimist
The question of whether g is some underlying neurological structure or if it's just a statistical artifact is a little more debatable than the question of whether it's anything at all; it clearly comes from something, and it's clearly predictive of relevant real life outcomes.

Originally posted by Astner
In order for a variable to be extracted from a mapping it has to be well-defined. General intelligence isn't, which is also why you can't measure it.

It doesn't have to be defined in the real world. If you have a dimension for each cognitive test, the graph will look like an ellipsoid and the longest axis is g.

It's a residual in an interpretation of a study that was done over a century ago.

Nope. It's one of the most replicated findings in psychology. I'll give more details when I'm less busy/maybe tonight.

Politics and activism are two areas that should definitely include the best and brightest. Our world would be a much better place.

Originally posted by The Ellimist
It doesn't have to be defined in the real world. If you have a dimension for each cognitive test, the graph will look like an ellipsoid and the longest axis is g.

So yeah, you're making shit up at this point.

Originally posted by The Ellimist
Nope. It's one of the most replicated findings in psychology. I'll give more details when I'm less busy/maybe tonight.

Just quickly before I have to go:

There's no point ever to orthogonalize statistical data, let alone test data.

Then why is it predictive? Why does a test of Raven's progressive matrices predict how good of a chief executive you'll be, and why does a test of verbal ability predict your ability to do higher mathematics? Why does g predict elementary tasks like reaction time?

Anyway, here's a good, semi-technical summary of the evidence, with rebuttals to the most common doubts (it's not as long as it appears - most of the length is user comments).

Originally posted by jaden101
2 and 3 typically advance 1.

4 seems to try it's best to hamper 1,2 &3

Originally posted by Robtard
Agreed, but wouldn't an influx of intelligence help alleviate th problems we face in politics?

Originally posted by The Ellimist
Then why is it predictive?

Originally posted by The Ellimist
Why does a test of Raven's progressive matrices predict how good of a chief executive you'll be, and why does a test of verbal ability predict your ability to do higher mathematics?

Originally posted by The Ellimist
Why does g predict elementary tasks like reaction time?

Originally posted by The Ellimist
Anyway, here's a good, semi-technical summary of the evidence, with rebuttals to the most common doubts (it's not as long as it appears - most of the length is user comments).

So you're going to dismiss the evidence and statistics provided because you saw people on the internet disagreeing with it? 😬

Originally posted by Badabing
On paper you would think. But aside from intelligent people we need incorruptible people.

Originally posted by The Ellimist
So you're going to dismiss the evidence and statistics provided because you saw people on the internet disagreeing with it?

There are no modern theories that relies on this particular value.

Originally posted by Lord Lucien
So we need Batman.

Basically.

Originally posted by Lord Lucien
So we need Batman.

We have far far too much activism already, we certainly don't need more. We need to get the activism out of academia.

The best and brightest should strive towards technological advances of all sorts. Advances in medicine, computers, sources of cleaner energy, etc.

The best and brightest is highly subjective here. What if one of the brightest people to walk the Earth, worked in psychology? How about social economics, or finding ways to re-educate criminals in a way that they were actually prepared to re-enter society, having gained tools that helped to prevent them from falling back into their old ways? Just saying. There are a crap ton of things that are needed in order to improve quality of life.

Originally posted by Astner

I'm really not interested in the debate between two bloggers

What? No, the necessary evidence is there in the opening post, which is a pretty short read. And it's not just the word of this blogger; plenty of academic papers are linked to for your convenience. The relevant points are:

1. There's a positive correlation between all test scores. This isn't something that was found in some random undergraduate study; it's ubiquitously found in every sufficiently large battery of tests that have something to do with an academic or cognitive task. Even people who try to come up with alternative theories of intelligence can't devise tests of their own models that don't end up with the g-factor.

2. You can statistically extract g from these batteries and figure out how "g-loaded" they are. The more cognitively demanding something is, the higher its g-loading tends to be. This is one of the most powerful cases against the sampling model (which is that the positive correlation is just the overlapping of sampling from various independent abilities); the g-loading of a test isn't very related to how similar in content type they are, .i.e. a hard math test correlates better with a hard verbal test than it would with a trivial math test.

(PS. it's only a perfect ellipsoid if the tests are perfectly g-loaded...)

3. g correlates even with elementary reaction time.

4. g predicts academic attainment incredibly strongly, and in most professions predicts job performance better than any single other measurable variable. It also predicts life expectancy, marriage success rates, etc.

5. Training for IQ tests, or any other sort of activity, produces improvements for that particular skill, but does not produce durable "transfer" to unrelated tasks.

6. IQ is very stable by adulthood.

7. Not sure if this part is in the link, but twin adoption studies and genetic analysis have put the heritability of IQ at as high as 0.8.

Now if you want to refute the data, you're welcome to. You can also try to show me any examples where you could construct a series of academic tests where the scores aren't correlated with one another. Literally nobody has been able to do this.

As a general note, you know that this stuff is widely accepted by consensus among psychometricians; it's hardly a fringe hypothesis.

I would included big data/analytics, though you could argue that technically towards the mathematics area I suppose since it has a basis in that. Business and other areas driven by that have more than shown their worth already and it's something that's really just begun being explored.