Taleb is wrong about IQ

http://www.jsmp.dk/posts/2019-06-16-talebiq/

by mpweiher • 7 years ago

150 264 7 years ago

chabes• 7 years ago

Tl;dr from the article:

All the claims from the article that I looked at, that can be interpreted as something specific and tested in a real data set, turned out not to be correct. If Taleb hadn’t blocked everyone who disagrees with him, perhaps he would have found out about this, and not published a post with all these incorrect claims.

sampo• 7 years ago

"The first principle is that you must not fool yourself — and you are the easiest person to fool."

https://en.wikiquote.org/wiki/Richard_Feynman

arethuza• 7 years ago

Feynman's IQ was only 125 - what would he know? ;-)

wwarner• 7 years ago

Wow https://www.psychologytoday.com/us/blog/finding-the-next-ein...

tw10104• 7 years ago

But Taleb is specifically arguing against simple linear correlation measures like R^2. (See figure 1 in Taleb's article.) Feels like you haven't grasped the specifics of his argument without taking that into consideration.

cheez• 7 years ago

Not only that but the best argument shows almost (but not quite) marginal contribution.

1000units• 7 years ago

Taleb's IQ is simply too high.

vipa123• 7 years ago

I don't understand. Are you saying the author shouldn't point out these errors, given what Taleb is arguing against? Or they aren't actually errors, given that the author hasn't grasped the specifics of Taleb's argument?

tw10104• 7 years ago

Taleb argues against R^2, and the author of this article seems to just ignore those points. I think Taleb has valid points, but the author here seems to be doing just the same conventional analysis that people writing psychology papers tend to do, which Taleb specifically argues is an invalid way of doing things (because assumptions built into linear correlation measures doesn't apply in this situation).

spamizbad• 7 years ago

This really needs to be the top post.

It's increasingly acknowledged, even in the "soft" sciences, that r-squared is not an appropriate analysis tool.

Anyway, wasn't the IQ test originally created to determine whether or not someone was mentally disabled? It seems like with more modern statistical methods it still could be useful measurement for that purpose, even if it doesn't accurately "scale" beyond 80 points. Ultimately, if you just want to answer the question "Does this person have significant cognitive deficits?" it still has its place even if it's not a good measure for broader intelligence.

reasonaway9• 7 years ago

but it does scale beyond 80 points. why ignore all the available evidence and trust instead the Taleb's evidence free assumptions?

zazaraka7• 7 years ago

Exactly. For example recently Taleb made a lot of fun of these linear fits, which are pretty ridiculous even for a non-statistician like me:

https://twitter.com/aedmans/status/1147494932183179264

zxcb1• 7 years ago

IQ is useful to me. Whenever someone thinks it's important I gain useful information about that individual

saas_sam• 7 years ago

Pretty cruel to ignore IQ and leave millions of mentally disadvantaged people in the lurch.

runarberg11• 7 years ago

You’re not wrong. IQ (or more generally intelligence tests) are very useful to diagnose people with some mental disabilities. That said: a) the IQ part of the intelligence test is sort of unnecessary for that purpose, and b) That’s usually not what people are (righfully) criticizing IQ for and you know it.

sremani• 7 years ago

That is Taleb's argument -- the IQ can be used to measure "incompetence" not "comptence".

An IQ less than say 85 means the inability of the individual and IQ of 140 does not mean that much.

His basic argument is, after certain IQ when you look at their economic outcomes they are evenly distributed.

hestipod• 7 years ago

The parent clearly meant the sort of people who use it as some status symbol. Some posturing of worth.

cwbrandsma• 7 years ago

I was useful to me as a learning disabled (dyslexic, ADHD, autism spectrum) individual, and when helping my kids in the same boat. But the extent to which it helped is just a slight confidence booster, "you can do this, you aren't dumb, even tho you are having issues in a couple subjects".

When you grow up failing spelling tests, under-perform in timed math tests and all sorts of random memorization subjects (things with lots of name and dates in them: biology, geology, etc), you get bogged down. Finding out my IQ, and the limits of the test, did help me understand myself a little bit better, and helped convinced me to get my college degree (my high school guidance counselor was telling me not to bother, just start working on the local farms, maybe get a little bit of community college).

sjg007• 7 years ago

Ouch.. those high school guidance counselors. I don't think an adult should ever limit the potential of a kid.

concinds4• 7 years ago

https://twitter.com/UntergrundmannG/status/11558354759463813...

"Taleb has an updated post up at medium that takes care of most of this. You could take the time to weigh that in against this all over again.

And if you come back here in a few days, I may have a post up about how this is a trivial misunderstanding of NNT's case. Quantitatively."

Anybody who understands Taleb's argument realizes that these "counterarguments" are missing the mark. The fact that they spend more time obsessing his "meanness to them" than on his arguments, without understanding that his meanness is never directed at people who disagree, only people who bullshit, proves that point.

jamesmadison662• 7 years ago

Completely misunderstanding NNT's arguments' basic premises in rebuttals to him seems pretty common.

pennaMan• 7 years ago

That's because he tends to go after things that are widely misunderstood in the first place.

mrandish• 7 years ago

Taleb is, as usual, a bit over-wrought and goes too far but this area has been a long-time interest of mine and I think he's directionally correct in that IQ is too often misapplied or misunderstood in practice. In general, laypeople tend to think IQ represents things it doesn't.

commandlinefan• 7 years ago

Being somebody who was always labelled as “smart” growing up - I was in the “gifted” program in elementary school, so there’s that - in a way, I want intelligence to be something that can be quickly, easily, and objectively measured, even if it turns out I'm actually not as smart as I’ve been led to believe I was. I can understand why this would make a lot of people uncomfortable, though. If IQ tests were ever shown to be perfectly correlated with that nebulous quality we label “intelligence”, it’s also not something you can improve. Your IQ, and likely your income level, are going to stick with you forever, just like your height will. Still, from what I can tell about this Taleb guy (other than that he seems like a difficult personality) is that he’s just trying to figure out how accurately IQ tests correlate with actual intelligence; he’s not railing against the concept of trying to measure it in the first place, as many others do.

js82• 7 years ago

I want intelligence to be something that can be quickly, easily, and objectively measured

I don't think it can be. I think (at least, important part of) intelligence is ability (or speed) to learn (patterns in the data?). So it can only be measured on something you haven't learned yet. Since as humans, we cannot "unlearn" things, the result will always be skewed by your previous experience.

frostburg• 7 years ago

You can improve your score by practice. It's one of the many issues.

mattnewton2• 7 years ago

I’m pretty unconvinced by this, the author makes a number of leaps that lost me. For instance, at the end of the article, the author says that Taleb’s claim that IQ only measures unintlelligence well isn’t true because at higher IQ, income continues to increase. This could be easily explained by numerous other third factors, like (hypothetically) IQ being a test for WASP culture and knowing WASP culture being correlated with better job opportunities.

liability2• 7 years ago

"like (hypothetically) IQ being a test for WASP culture and knowing WASP culture being correlated with better job opportunities."

These sort of arguments always ring hollow for me, because I am not a WASP and the test I took didn't resemble a WASP-culture test in the slightest. The test I took was all "given this series of abstract geometric patterns, which comes next," not, "During which season is the weather at Martha's Vineyard the most temperate"

Maybe wrapping your mind around abstract geometry is something WASPs are good at and, say, people from the Islamic world suck at, but I really really doubt that considering the relatively elevated roll geometric patterns have in Islamic culture compared to WASP culture.

Of course I am neither a Muslim nor a WASP, but even so I am pretty sure I didn't struggle too hard on that test.

mattnewton2• 7 years ago

That’s what the “hypothetically” was for- it was meant as an illustration that there could easily be a third effect loosely correlated with both. For this to dispute Taleb’s claim, you have to think that income is also correlated with intelligence and not more strongly correlated with other things that IQ could be measuring.

That example probably popped into my head first, because of the long controversy around IQ tests cultural effects and the history of using those tests to make arguments that race determines intelligence

Consultant32452• 7 years ago

There's a lot of ways to eliminate cultural bias in IQ testing. For example, you can just take a set of people that are not WASPs. You will find their relative outcomes still correlate strongly with IQ relative to their peers in the non-WASP culture.

Even if you feel that Psychologists have failed to do so, you still have to explain why East Asians and Ashkenazi Jews consistently outperform WASPs.

reasonaway9• 7 years ago

IQ predicts more than just income. It has incremental value for predicting scientific publications, academic achievement, patents etc even within the top percentile. http://adamlgreen.com/smpy/

there is also no evidence of strong cultural bias of the sort you are speculating about.

b1gtuna4• 7 years ago

Off topic, but how does one go about to measure his/her IQ?

zazaraka7• 7 years ago

You can find free online tests. They should take at least 30 min. Do 3-4 different ones to confirm you get about the same score (less than 5 points difference).

If you live in a big city, you can probably find a place to do it for less than $100.

b1gtuna4• 7 years ago

It looks like there is general distrust on the online tests (searching Reddit posts). Can you recommend a few if you know any?

dgellow• 7 years ago

A lot of companies provide that kind of thing, you can easily find (good and not that good) IQ tests by searching for "standardized IQ tests" (or the equivalent in your local language). You can also contact the Mensa organization and they can put you in contact with a testing coordinator (see https://www.mensa.org/mensa/about-us).

b1gtuna4• 7 years ago

Thanks for the link. I searched the internet, but I couldn't figure out which ones are the good ones.

b1gtuna4• 7 years ago

FWIW, I googled IQ Test and blindly clicked on the first link that appeared. Ran through 30 questions and when I clicked on submit - it prompted for a payment lol. Should've seen this coming!

sunseb3• 7 years ago

This is the elephant in the room that explains why IQ is such a controversial topic:

https://brainstats.com/average-iq-by-country.html

https://www.abc.net.au/news/2019-01-14/nobel-prize-winner-ja...

zazaraka7• 7 years ago

There does seem to be a mild correlation with outside temperature. Could some places just be "too hot to think"?

fastball16• 7 years ago

Australia/SE Asia is very hot, as is Mexico/SA.

defertoreptar• 7 years ago

I wonder if it could be the difference between areas that rely more on agriculture vs areas that must rely on their tertiary sector (where education counts for more).

throwaway123x2• 7 years ago

I would correlate it more with food security than with heat.

wsc981• 7 years ago

There does seem to be a mild correlation with outside temperature. Could some places just be "too hot to think"?

Well ... perhaps in some way:

    Since 1991 there has been the presumption that persistent heat affects the IQ over 
    generations. It was initially assumed that living in lower temperatures requires a higher
    physical fitness and causes higher social demands. Richard Lynn, a professor at the 
    University of Ulster in Northern Ireland, also assumes that the colder temperatures 
    increase the brain volume. Whether brain size is at all related to intelligence, is 
    controversial.

    In general, a hot climate is considered a disadvantage for the development of 
    intelligence. The reason is nowadays primarily seen in the high physical energy demand 
    and the resulting stress.

Source: https://www.worlddata.info/iq-by-country.php

Another study suggests infectious diseases might be a cause for lower IQ scores. In that case, perhaps infectious diseases spread more easily in hot climates?

See: https://www.scientificamerican.com/article/why-is-average-iq...

tw10104• 7 years ago

We can keep playing the "maybe there are other factors besides DNA that could explain small differences in IQ"-game until the cow comes home. The sad thing is that eventually we'll probably find some genuine correlation between genes and some measure of intelligence (controlling for every other conceivable factor). (Maybe we won't, but I wouldn't bet on it in a prediction market.) Not sure what'll happen once that day comes.

sunseb3• 7 years ago

There is this theory that life in the Northern Hemisphere is very hard (because of Winter), so there may have been some kind of a harder natural selection there in the past. People in the North seem to be obsessed with building, working and thinking about the future ("I need to be ready, winter is coming").

octohedron3• 7 years ago

Singapore is first in the world and it's right in the equator, while Hong Kong, the other one at the top has a humid subtropical climate. Are you perhaps from a cold area? That would easily explain your bias.

EForEndeavour6• 7 years ago

From a glance through the list, there's also a decent correlation with socioeconomic development, which is to be expected. There's a ton of multicollinearity going on.

CBLT• 7 years ago

This was the accepted scientific understanding circa 1900. Interestingly enough, this was used by early activists to argue that global warming is a real phenomenon. The logic went: since people indigenous to tropics cannot have advanced civilization, and Angkor Wat is clearly a product of advanced civilization, then Cambodia wasn't tropical when it was built.[0]

[0] https://newleftreview.org/issues/II97/articles/mike-davis-th...

fastball16• 7 years ago

Yeah, I'm honestly not sure how to interpret this.

Either you assume that IQ is 100% not genetic, or you have to admit that race seems to be correlated.

js82• 7 years ago

I think the study is controversial because there is huge bias in the way people are educated in these different countries.

No human capability is likely to be 100% genetic.

marliechiller• 7 years ago

theres an interesting sam harris podcast featuring charles murray on this topic which is well worth the listen imo: https://www.youtube.com/watch?v=S6Rmbfi6YrE

Barrin92• 7 years ago

the correlation isn't the controversial part, the fact that IQ is one, or the dominant causal driver for different outcomes is.

and also, there is not a single piece of evidence on this planet that IQ is 100% hereditary.

Skunkleton2• 7 years ago

Either you assume that IQ is 100% not genetic, or you have to admit that race seems to be correlated.

Or, bear with me here, you have to admit that IQ tests can only measure trained skills that correlate loosely to raw intelligence.

BurningFrog2• 7 years ago

If IQ isn't genetic, rabbits would have the same average IQ as humans.

But that does not mean "races" have to have different inherent average IQs.

root_axis• 7 years ago

Of course there are racial correlations. The "races" are 5 gigantic buckets that abstract over an unimaginably complex variety of DNA. The reason why people get so worked up over this is because someone always chimes in with "based on these correlations we need to face the hard truth that black people are dumb and ashkenazis are geniuses, act accordingly".

ajross6• 7 years ago

So, here's a reasonably rational response to a hotbutton claim like that: I think most numerate folks would look at that data and view a "race/IQ correlation" as at least a reasonable hypothesis.

But come on, it's weak. The relationship that chart shows clear as day is wealth. Wealth is super correlated. The top of the chart are exlusively industrialized democracies. The bottom are exclusively impoverished (and mostly unstable) backwaters.

And within those cohorts, sure, you can kinda/sorta say that east asians sit above europeans, maybe. Or maybe not. Four points of IQ are nothing -- it's like three decades of development per the Flynn effect, or maybe one decade of really focused investment in education.

There are counteless other good hypotheses for a weak signal, and this isn't an area amenable to direct experiment, so realistically we're never really going to know.

And the same stuff is true of the linked article. It's poking holes in specific statements by finding rigorous numbers that "disprove" it, but only very weakly. Basically, the point to the article isn't that "IQ is solid science", it's that Taleb's paper was too aggressive in its conclusion. Which is sort of a yawner, and not really a good excuse to bring race into the discussion.

But yeah. It's possible that darkies are dumber or that the gooks are smarter. It's possible. What drives those of us on the left absolutely bonkers isn't the possibility that you might be right, it's the fact that you look at messy data like this and immediately go to the least tolerable conclusion as if that's what you want to be true. And... yeah, yuck.

(Edit: and right on cue three quick responses about "but that doesn't prove race isn't correlated, here's another hypothesis for why race might be correlated!", which sounds to me suspiciously like "I really want race to be correlated with IQ!" The truth is we just don't know, we might never, even if it is it's weak, and I for one would rather live in a world where people don't want this to be true than the one I'm stuck in. )

RobertoG• 7 years ago

That's not the only available explanations. You can find many alternatives. Specially now that we have discovered the influence of epigenetics.

For instance, average height of a population is not a fixed value but changes with the progress of a country. Presumably due to the diet improvement in the mother's generation.

Another consideration: parasites, I would be surprised if they don't influence the development of child intelligence.

openasocket2• 7 years ago

Two counterpoints:

First is that there are plenty of environmental factors to consider, especially with regards to 3rd world countries. In rural parts of Africa where people are predominantly subsistence farmers, there is a very high rate of prenatal and neonatal malnutrition. And that can cause mental retardation. And so can getting malaria and a variety of other infectious diseases in infancy.

Second point is that you are assuming that race is a well-defined genetic construct. But it isn't, it's largely a social construct. Population genetics for humans is a continuum, and we just arbitrarily draw the lines.

bitcurious2• 7 years ago

There’s isn’t anything controversial with admitting that race is correlated with IQ.

The controversy is from folks trying to take that correlation, reframe it as causation, and then attribute it to genetic factors.

There are tons of non-genetic factors that can affect IQ and are correlated with race: exposure to pollution, economic situation, nutrition, etc.

Then there is the test itself. I’m not sure how localization happens, but it’s a pretty hard problem that lends itself to biases.

kilo_bravo_3• 7 years ago

I assume that IQ is correlated to poverty, educational attainment, the learning styles used in schools, and the presence or lack thereof of "cram" schools.

The top 5 highest scoring countries, Hong Kong, Singapore, South Korea, Japan, and China all have a cram school culture that prepares them for the endless rounds of standardized tests they have to take to secure a position in society.

I would expect people trained from infancy to excel at standardized tests, to excel at standardized tests.

Many of the reasons I listed would explain why, in the US, the states of West Virginia (98.7) and New Hampshire (104.2), despite having very similar demographics have such different average IQs. One state is richer than the other.

It would also explain why Hawaii, which demographically is comprised of people (about 1/3rd) who share genetic lineage with people from high IQ nations, has a relatively low average IQ. Hawaii has a pretty bad public school system.

Until someone locks two children in an Edu-Pod from birth to adulthood and feeds them the same stream of information and nutrition and measures their IQ when the pod ejects them at the end of the test, we'll never know for sure, but my gut tells me that all of the people who think "genetics is destiny" are losers.

concinds4• 7 years ago

The fact that the "IQ crowd" often falls back to racial arguments like this (based on shoddy statistics) shows why people should be deeply suspicious of their motives.

"We need to openly talk about race differences in IQ, lest the neo-nazis appropriate it and use it as recruitment propaganda" is a very common refrain, and pretty transparent in its intentions, as it itself acts as recruitment propaganda.

https://twitter.com/nntaleb/status/1116342266195476481 <- this Twitter thread addresses the country-IQ argument effectively, and I haven't seen it properly refuted so far.

mcguire12• 7 years ago

If one ever finds oneself in the position of invoking a book with the title, Coming Apart: The State of White America, 1960–2010, in support of one's arguments, one may wish to reconsider how one is living one's life.

ceejayoz4• 7 years ago

The elephant in the room for the "it's all about race!" folks is the Flynn effect, though. https://en.wikipedia.org/wiki/Flynn_effect

selimthegrim3• 7 years ago

Why isn’t Sierra Leone colored correctly on that map?

yters3• 7 years ago

It'd be interesting to control for their education environment. My understanding is many of the Asian countries ranked at the top are almost constant schooling for children up through early adulthood, which is not the case for the US and especially not the case for many countries farther down the list.

The schools tend to be very heavy in STEM subjects, which does not seem an implausible source for high IQ test performance, which favor the kind of abstract, logical thinking that STEM subjects strengthen.

So, IQ could be a measure of abstract thinking capability, and it might be the case that one's ability to think abstractly can be developed through proper education.

fastball16• 7 years ago

That doesn't seem to be the case, as IQ performance does not tend to get any better for an individual after extensive education.

itsgrimetime• 7 years ago

You're right, IQ has been used to harm groups of people (specifically black and brown people) for a long time. If I didn't know you were just trying to post reactionary propaganda I'd say I was surprised you decided to leave out all the data that shows that IQ doesn't encompass all ways a person can be "intelligent".

"The assertion that IQ measures human intelligence in any general sense, or that the source of variance in IQ scores is primarily cognitive in nature, remains unsubstantiated after decades of investigation.": https://pdfs.semanticscholar.org/0382/3398d781543cd0edcf51f1...

https://www.cell.com/neuron/fulltext/S0896-6273(12)00584-3

If you're going to try to convince people of your "race realism" bullshit on this platform you're gonna have to try a lot harder than unsubstantiated shit posting.

mcguire12• 7 years ago

Taleb: "If you looked at Northern Europe from Ancient Babylon/Ancient Med/Egypt, you would have written the inhabitants off as losers who are devoid of potential... Then look at what happened after 1600. Be careful when you discuss populations."

A side note: Taleb's writing ability goes to hell when he rants.

zajio1am• 7 years ago

The most fascinating thing for me in the article is bimodal distributions in the last grap, where lower peak is ~same as the peak for low-iq distribution. I wonder what is the cause for this.

jedberg3• 7 years ago

The problem is, it has been show time and again that most IQ tests are socio-economically biased. And every "proof" in this paper is a correlation that would also hold true is you replaced "high IQ" with "had rich parents".

tw10104• 7 years ago

That's not true. There are definitely papers that tries to control for socioeconomic factors.

mc324• 7 years ago

Studies with twins could offer some form of either repudiation or confirmation.

skybrian3• 7 years ago

Well, the key word here is "try". There is evidence suggesting that controlling for confounding factors is very difficult and many studies don't succeed. From the outside, I'm not sure how much confidence we can have that they succeeded?

https://slatestarcodex.com/2019/06/24/you-need-more-confound...

ctvo• 7 years ago

I don't know enough about IQ to refute this, but let's assume it's correct. Taleb would still be wrong about his IQ claims as is the point of this post.

mc324• 7 years ago

I think that would be better represented as “has access to quality education and gets trained in areas needed by the economy”.

seieste• 7 years ago

How is Raven's matrices biased?

Hermel2• 7 years ago

Have you considered the option of IQ being partially inherited and the parents also being rich because of a high IQ. In that case, IQ would still be the cause of the wealth.

jedberg3• 7 years ago

Intelligence has been proven to be partially heritable. I don’t think that’s in question.

But the claim is that IQ measure wealth, not intelligence.

fvdessen4• 7 years ago

Well, yes since IQ is the largest predictor we have for socio-economic achievement, and is highly heritable. Rich parents have rich kids & smart parents have smart kids & more often than not the two happen at the same time, because being smart is the best predictor of financial success.

jedberg3• 7 years ago

That’s a circular argument though. If the IQ test is biased towards rich people then of course rich people will score better.

freetobesmart• 7 years ago

Im curious what people will get out of determining measurable IQ. What are their end goals.

1) You dont get smarter. Its not like knowing you are 120 vs 100 IQ changes anything about your ability to solve a problem

2) IQ is more of a speed of learning not an ability to understand. With enough time you can learn anything (challenge placed) the biggest barrier is discipline and desire.

3) IQ is abstract and is very useless within any context. If you want to know if someone is good at math give them a math test. If they are a good writer look at their portfolio. IQ is like determining someone is good at being a chef by only asking them questions. IQ is very far from any specific field of knowledge.

IQ is like space knowledge. Its only important/useful subject for a select few people.

ohaideredevs6• 7 years ago

1. You may want to choose a less intellectually rigorous career if your IQ is low. Or you might get an early warning to work on factors you can control - for example the ability to focus, getting enough sleep, etc.

2. "With enough time..." is a totally irrelevant argument made by people who don't understand competition and how it applies to the real world, as in for scholarships, promotions, etc. Time is everything - and its a luxury. It also cuts into your free time heavily.

3. This is true. Except, when you read some code or a complex math concept, have fun if your working memory is sub-par and you can't even remember what the inputs to a function are.

"Its only important/useful subject for a select few people."

It's the exact opposite of that - it is applied to learning any field fast.

nshepperd6• 7 years ago

IQ was important to determine the effect of lead exposure on cognitive function. Without it we probably wouldn't have nearly as stringent laws banning it (eg. in paint, and gasoline), and would still be suffering from endemic lead poisoning.

It's not particularly important to know your own IQ as an individual (you probably already have an idea of your abilities) but it's very useful for that sort of public health purpose. If there are other neurotoxins floating around which we don't know about, IQ will be important to identify and justify banning them.

https://www.ncbi.nlm.nih.gov/pubmed/8162884

tompccs3• 7 years ago

This does not address the substance of Taleb's arguments. Read Taleb's original article:

https://medium.com/incerto/iq-is-largely-a-pseudoscientific-...

He's showing us that psychologists' statistical methods are faulty, that IQ advocates don't understand the underlying mathematics and blindly plug things into R. Lo and behold, in rebuttal, psychologists try to prove Taleb wrong by plugging more things into R.

Radle• 7 years ago

Doesn't mean the psychologists are wrong.

tompccs3• 7 years ago

Well, IQ is their measure. They should be able to defend it properly.

And frankly, if there is any doubt that IQ maps to intelligence, then its abuse over the last half-century has done immesurable harm to individuals and given considerable fuel to racists.

moomin• 7 years ago

Yeah, a bust clock is right twice a day. Maybe that's one of those times.

cljs-js-eval5• 7 years ago

No, but these [1][2][3] sure do:

[1] https://journals.sagepub.com/doi/full/10.1177/17456916156099...

[2] https://osf.io/ezcuj/wiki/home/

[3] https://science.sciencemag.org/content/349/6251/aac4716.full...

You should have very little confidence in the predictive power of most psychology findings since the founding of the field.

devoply4• 7 years ago

Talebs' main beef is that the system selects for drones that serve the system and can think on its terms and the question is obviously why would it not do that. All systems protect their own integrity and obviously the IQ test would further serve to protect the power structure of the people who designed those tests. And the effects of selecting for these sorts of people is obvious everywhere in a dysfunctional system which pretends to care about critical thought and different ways of thinking only to subvert those and use those to increase its own control and power... despite not being a good candidate for managing the affairs of the planet.

ohaideredevs6• 7 years ago

Raw capability has nothing to do with how that capability is applied. IQ is supposed to measure raw capability. Especially tests such as Raven's which focus on fluid intelligence.

macspoofing5• 7 years ago

All systems protect their own integrity

Do they? Where are you getting that from?

and obviously the IQ test would further serve to protect the power structure of the people who designed those tests.

Not so obviously. No matter what IQ is, it was neither designed nor implemented as a way to 'protect the power structure'. This idiotic power hierarchy metaphor is completely overused by certain ideological sectors. Not everything is best understood as power hierarchy. Not everything is a power hierarchy. When you claim everything is a power hierarchy it is a cognitive crutch not unlike conspiracy theorists seeing the work of the illuminati behind every socio-political event.

nabla9• 7 years ago

Generic arguments about bad methodology don't invalidate specific arguments about IQ. If he used bad statistical arguments himself in this case, we are in impasse.

Of course Taleb understand statistics if he wants to, but he is just pushing his rants and loses his bearings in Twitter. Even his old friends are lamenting his online personality and bitterness.

iQuercus• 7 years ago

What does his personality have to do with what he is arguing? Is he using “bad statistical” arguments or not?

ilikehurdles• 7 years ago

Of course Taleb understand statistics if he wants to, but he is just pushing his rants and loses his bearings in Twitter. Even his old friends are lamenting his online personality and bitterness.

Generic arguments about personality and character don't invalidate specific arguments about IQ.

If I indulge your "argument", from what I see on twitter is threads full of very openly nordic-/white- supremacist keyboard warriors and zerohedge trolls. Of all the people to call out in these threads, Taleb seems the most level-headed.

cljs-js-eval5• 7 years ago

I really, really wish people took bad methodology criticisms more seriously. I have a degree in the social sciences, and if there is one thing that completely defines the field right now it is that you can't use their findings for absolutely anything.

A good portion of studies don't replicate, including fundamental ones (in particular see the failed replication here [1] on one on Rand et al, a study on the effects of priming). Priming is a huge topic in social psychology.

He understands statistics. I'm not sure his opponents properly understand that when you complain about methodology, the implications are not just "add a section in the paper acknowledging the potential for error", it's "your whole study might be fatally flawed in a way that invalidates all your conclusions".

[1] https://authors.library.caltech.edu/91063/2/41562_2018_399_M...

cjhanks3• 7 years ago

I love in that article the complete non-sequitur about MENSA members wearing white socks with birkenstocks. It really demonstrates his disdain.

mcguire12• 7 years ago

Ignore the rant. Look at the graphs.

zazaraka7• 7 years ago

Those MENSA members were truly fashion trend setters, as expected. The rest of world only now realizes how cool white socks with birkenstocks really are:

https://hypebae.com/2018/11/how-to-wear-birkenstock-sandals-...

reasonaway9• 7 years ago

he examines specific claims made by Taleb in detail and shows them to be completely wrong in every case. that can be reasonably described as addressing the substance of his arguments.

can you identify any specific problems with the analysis in that post?

discussions around taleb often have a very cult like aspect to them - a refusal to discuss specific empirical claims made by Taleb and an assumption that the critics are too dumb to understand taleb's deep mathematical criticisms.

mpweiher• 7 years ago

Article was specifically not looking at the non-quantitative ranting: "Most of the document is just a rant, but he does make some quantitative claims, which is good. I investigate the veracity of these claims using real data."

So he looks at the quantitative claims and they are just flat out wrong.

However, the non-quantitative "analysis" is just as faulty.

First huge mistake is that he claims that IQ correlating better with outcomes at the low end somehow shows that it doesn't measure what it claims to measure, "raw ability". That is nonsense, because ability is just one factor, but a necessary one, for outcomes. So if you don't have the ability, you're not going to get the outcome. But if you do have the ability, that doesn't mean you're going to get the outcome. You might not be motivated, for example.

In fact, I remember a talk from a couple of years back showing that financial success tends to be anti-correlated with the highest IQs, because those people figure out that it just isn't worth it.

But anyway, the correlation being stronger inn the "non-ability" part of the distribution does not show that it's not measuring ability.

Another whopper is that he claims that performance is is fat-tailed, without citing any evidence. According to Albert-László Barabási[1] who studies performance, success and how they relate, performance is more closely normally distributed, it is success that is fat-tailed.

And of course Taleb doesn't seem to distinguish between performance and success, which makes the whole rant pretty worthless.

[1] https://en.wikipedia.org/wiki/Albert-László_Barabási

Will_Parker• 7 years ago

Some things become clearer once you understand: a lot of people who say "X is invalid" really mean "We are better off not talking about X".

dekhn• 7 years ago

IQ arguments are the most boring thing I've seen hackers engage in. Even worse than blockchain and genetic engineering in terms of poor scientific quality, immediate jump-to-conclude-my-personal-pet-theory-is-right-because-anecdote, ignoring social factors.

macspoofing5• 7 years ago

They are boring, and I'm not even sure what the argument is. On the edges, I think there's a general agreement that low IQ is correlated with mental disability, and on the upper end high IQ is correlated with a certain type of cognition that makes you a good professional Mathematician or Physicist. I'm not sure what the big middle really says about anything.

If I were to guess why this IQ debate keeps coming up is that it is a bit taboo now (which makes it very tantalizing in and of itself). There is also some gaslighting going on with certain segments claiming IQ is completely junk and has no genetic basis (even partially). This triggers the contrarians, who then themselves may overstate the science, prompting responses that may understate the science ... and so on.

mistrial9• 7 years ago

https://en.wikipedia.org/wiki/Theory_of_multiple_intelligenc...

there are versions of this, but is anyone surprised that an attempt of big-S Science to put a single number on humans, has fundamental challenges ?

cljs-js-eval5• 7 years ago

One of his repeated criticisms of analysis like this is using regression to fit datasets that don't necessarily follow a straight trend line.

His larger point with respect to IQ is that it only predicts substandard IQs. Measures like the variance are largely influenced by the effects of IQs below the average, not above the average.

A second point which I don't see addressed is that IQ measures share features with testing. If how you test in school affects future income, and testing well in school predicts testing well on IQ, then more robust analysis is needed. A sample space of people in a wide band of IQ ranges who all got about the same grades in school would be a true test of the predictive value of IQ on income.

reasonaway9• 7 years ago

it is all unfounded speculation based on mathematica simulations and his own assertions. there is no empirical support for his arguments.

tomp• 7 years ago

One of my favourite papers discusses this (and more):

Top 10 Replicated Findings from Behavioral Genetics
1. All psychological traits show significant and substantial genetic influence
2. No traits are 100% heritable
3. Heritability is caused by many genes of small effect
4. Phenotypic correlations between psychological traits show significant and substantial genetic mediation
5. The heritability of intelligence increases throughout development (from 41% (age 9) to 55% (age 12) and to 66% (age 17); "unlike the other findings, this one is limited to a specific domain")
6. Age-to-age stability is mainly due to genetics
7. Most measures of the ‘environment’ show significant genetic influence
8. Most associations between environmental measures and psychological traits are significantly mediated genetically
9. Most environmental effects are not shared by children growing up in the same family

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4739500/

yters3• 7 years ago

The 2000 SAT is supposedly an IQ test, and I boosted my score by 100 points by doing 10 practice exams. So, IQ probably tests some objective human faculty, but insofar as the 2000 SAT is an IQ test, it is clearly the case the IQ tests can be studied for (hence all the expensive SAT prep tutors).

squirrelicus6• 7 years ago

Can we please agree that recognizing the rigorous science on the heritability of IQ does not imply racism?

olliewagner• 7 years ago

I believe his biggest issue with IQ—one that worries me greatly as well—is that it is being used to create a feeling of "otherness" towards populations that can be demarcated out using a single dimensional property: IQ. History shows us this as a strategy to subjugate using other "scientific" methods of classifying people, phrenology being a fairly recent example.

mcguire12• 7 years ago

Taleb: "“IQ” is a stale test meant to measure mental capacity but in fact [measures] to a lesser extent (with a lot of noise), a form of intelligence, stripped of 2nd order effects — how good someone is at taking some type of exams designed by unsophisticated nerds."

Well, that statement is clearly false. The exams are designed by very sophisticated nerds.

arendtio• 7 years ago

Actually, I don't care much about if the IQ can measure the mental capacity precisely or just find those who are mentally disabled. I think the core problem is, that our society thinks of an IQ as some magical figure to determine the value of a human being.

I think an IQ is much closer to what the clock rate is for computers. Just because a computer runs at a higher frequency, doesn't mean that it is computing more important calculations.

In fact, there is no advantage in having 3 times the processing power if you are just running a word processor. Equally, there is no advantage in having 3 times the IQ if you are just going to shop groceries.

Yes, there are a few use-cases where having more processing power is relevant (e.g. when you have a limited time to make a good and impactful choice), but I think those cases are not a frequent as most people think they are.

dennis_jeeves• 7 years ago

Analogy (sarcasm):

-Height differences do not exist because the measuring scale used to measure height is inaccurate.

-Height differences are not dependent on hereditary because height depends on nutrition.

-Assuming that height differences between the races are inherent tantamounts to racism because height differences are clearly depend on nutrition.

-Bad nutrition is caused by discrimination, since bad nutrition results in shorter heights it's clearly discrimination that indirectly causes shorter height.

-Height is not hereditary because short parents can have tall children and vice-versa.

-Person X measures very short using a 'height' scale, but he's a great basketball player. Hence obviously height is not a factor in being a great basket ball player.

Taleb is wrong about IQ

Related Stories