this post was submitted on 22 Nov 2024
1120 points (98.9% liked)
Science Memes
11217 readers
2651 users here now
Welcome to c/science_memes @ Mander.xyz!
A place for majestic STEMLORD peacocking, as well as memes about the realities of working in a lab.
Rules
- Don't throw mud. Behave like an intellectual and remember the human.
- Keep it rooted (on topic).
- No spam.
- Infographics welcome, get schooled.
This is a science community. We use the Dawkins definition of meme.
Research Committee
Other Mander Communities
Science and Research
Biology and Life Sciences
- [email protected]
- [email protected]
- [email protected]
- [email protected]
- [email protected]
- [email protected]
- [email protected]
- [email protected]
- [email protected]
- [email protected]
- [email protected]
- [email protected]
- [email protected]
- [email protected]
- [email protected]
- [email protected]
- [email protected]
- [email protected]
- [email protected]
- [email protected]
- [email protected]
- [email protected]
- [email protected]
- [email protected]
- !reptiles and [email protected]
Physical Sciences
- [email protected]
- [email protected]
- [email protected]
- [email protected]
- [email protected]
- [email protected]
- [email protected]
- [email protected]
- [email protected]
Humanities and Social Sciences
Practical and Applied Sciences
- !exercise-and [email protected]
- [email protected]
- !self [email protected]
- [email protected]
- [email protected]
- [email protected]
Memes
Miscellaneous
founded 2 years ago
MODERATORS
you are viewing a single comment's thread
view the rest of the comments
view the rest of the comments
I understand that you're saying there are more incredible geniuses than full on retards.
However, IQ scores are a normal distribution with an arbitrarily defined mean of 100 and a standard deviation of 15.
So, IQ scores of 0 or 200 are both 6.6 standard deviations from the mean. If IQ is truly a normal distribution, you'd expect the number of people with IQ scores <= 0 and the number with scores >= 200 to be exactly the same, simply because this is how the scores are defined.
If you try to look up what proportion of the population falls outside 6.6 standard deviations, the z-tables don't go out this far. It's essentially 0% (0/100) but how many is it out of 8 billion?
It's not. Here's a list containing a number of people above 200.
However, no-one has a negative IQ.
any article that lists historical figures with even estimates their IQs can be discarded as bullshit. IQ has specific testing criteria and imo the most important part of it is its basis in general distribution - if we don't know the IQ of the average peasant, we can't know the IQ of Shakespeare
besides, IQ is a borderline pseudo science to begin with. i was made to take an official IQ tests and the second i stepped out of the test room i started wondering how is this going to accuratly portray my "innate" intelligence when the vast majority of the things on the test can be learnt or otherwise trained to be better at
There are people alive on that list.
The person above is trying to prove IQ legitimacy with normal distributions and confidence levels. I'm arguing against it.
i have indeed noticed there are people alive on that list. But are you going to trust a source that states someone's IQ to be literally outside of the possible scale when it also just makes shit up a few people down?
i don't think they're trying to prove IQ's legitimacy, just explain the way it's calculated
Let's focus on one individual then with an officially calculated IQ.
https://medium.com/@gigasociety/younghoon-kim-the-current-highest-iq-276-record-holder-in-2024-65d73e5a88c5
IQ is not normally distributed. It can be higher than 200. It can't be negative.
https://www.quora.com/Is-it-even-possible-for-a-human-to-have-an-IQ-of-200?top_ans=179514973
read the second answer to that particular quora question, i believe it outlays what the other guy and i mean pretty clearly
This is repeating the same confusion.
Calculating values from the normal distribution tells you nothing about the tail properties of human intelligence.
yes, just as IQ tells you close to nothing about the properties of human intelligence! (only how good you are at taking IQ tests).
Consider this - what does being smart even mean? Does it mean you're able to solve logic puzzles fast? Does it mean having a good memory? Does it mean being able to make good decisions? maybe it means being able to resolve interpersonal conflicts? or maybe being able to cook something amazing from scratch without a recepie?
IQ seems to be seen as some vague concept of the computational power of the brain, but only when it comes to logic puzzles and remembering things. What if someone's brain's computational power instead favours considering the interactions of various flavours to create outstanding dishes? or moving their body to dance the most mesmerising dances?
imagine you're a scientist though! a man of science, logic and reason, living roughly at the same time IQ was standardised. And you are smart, all your friends think you're smart - so you set the scale of the entirety of human intelligence to be measured with logic puzzles. Nothing else. All the other stuff is just some talent someone has...
but what if someone is talented at solving IQ tests? Does that mean they're smart? if there is no discernible difference between someone who's talented at solving logic puzzles, and someone who an IQ tests deems to be intelligent, does that mean only those who enjoy logic puzzles, and therefore have gotten a lot of practice in solving them, are smart?
another question - is it "cheating" if somebody trains for their IQ test? if someone trains their mind specifically to be better at them - will that person become more intelligent, or just more skilled at filling out IQ tests well? how can you spot a "cheater" like that?
where even is intelligence in the brain? where does it come from? your genetics? your upbringing? your environment? everything at once?
how do you measure something you can barely define? and why with logic puzzles? why not an interpretative dance to the sounds of noise jazz? why not the baking of a pavlova cake? or maybe a rap battle?
apologies for the long rant. IQ is not a scientific measurement, it's a measurement of how likely you're to do well on logic puzzles. and whoever popularised it and made it seem like the way to prove you're better than others infuriates me. the above are my personal, more or less subjective, issues with the idea of IQ, i do recommend this video essay to understand how deeply flawed even the history of IQ is. There's piles and piles of arguments against IQ, and very few in favour
Agree with all of that. What annoys me is when properties of the normal distribution are used as "facts" about human intelligence.
Reread my original statement and see if you still disagree
eh, well my answer is going to be most likely unsatisfying because - that just depends on how you count it, there's quite a few different IQ tests and some of them use slightly different methods of calculating the scores
practically though? a person so disabled they can barely figure out the most basic puzzles that scores below ~20 would probably have significantly lower survival chances basically anywhere, but especially in developing countries where they're less likely to get help
I have to disagree.
IQ as a measure of intelligence doesn't work that way. The number can't just get higher and higher because a person is really smart. A supreme, godlike intelligence doesn't have an IQ of say, a million.
IQ has a statistical definition and although intelligence may not follow a perfect normal distribution, IQ Score does.
If there are about 8 billion humans, then 1 of them is "the smartest" in some way. 1/8,000,000,000 is 1.2x10^-10, this has a z score of 6.33.
The current smartest person will have an IQ of (6.33x15)+100=195. No one has an IQ of 200. This isn't because a person can't be any smarter, it's because this is how IQ is defined. If a pure, perfect, godlike intelligence exists in our current human population, their IQ is 195.
I linked to a list of many examples
Only if normal distributions are assumed. Clearly this assumption is incorrect.
But we do agree that a negative IQ is impossible?
You provided a link to reader's digest. It's not the most credible reference.
A negative IQ score and an IQ score above 200 would be possible with larger populations.
A 200+ IQ is possible with a small population. Normal distributions are not a physical law.
I'm struggling to see how a negative IQ can be practically assessed.
I think the confusion is that IQ is not an objective measurement. It's subjective.
Its not like say, height, where you can have a normal distribution and then a statistical outlier.
The IQ point isnt a constant, tangeable unit of measure, like an inch. Intelligence isn't something you can put a ruler up to and say, oh that's weird, this person has an IQ of 300 and is a statistical outlier.
IQ is defined statistically. You use some method of claiming that each person has a certain ranking of intelligence. Then you use a defined mean and SD to determine what IQ value that corresponds to, in the context of everyone else in the population.
Yes, a ranking. Ideally the same test for the whole population.
Here is your error. Limiting the description of the population distribution to only 2 parameters severely restricts the range of distributions that can be selected. Forcing the population distribution to be Normal is done for arithmetic convenience only. Not because intelligence must be normally distributed.
I'm not saying intelligence is a normal distribution. I'm saying that IQ scores are a normal distribution.
The metric, IQ is a normal distribution because that's how the metric is defined.
I'd like to hear your explanation how an IQ of above 200 is possible and what that would actually mean.
Its only possible if there are about 10x more humans. With a population of around 80 billion, the smartest one person would have a z score of roughly 6.6 and an IQ of roughly 200. This is calculated from a mean of 100 and a standard deviation of 15, which is how it's defined.
Here's a reference from Wikipedia for you, which, itself, references many scientific journals:
" IQ scales are ordinally scaled.[81][82][83][84][85] The raw score of the norming sample is usually (rank order) transformed to a normal distribution with mean 100 and standard deviation 15.[3] While one standard deviation is 15 points, and two SDs are 30 points, and so on, this does not imply that mental ability is linearly related to IQ, such that IQ 50 would mean half the cognitive ability of IQ 100. In particular, IQ points are not percentage points "
So, as I've been saying, you just put everyone's test scores in order from worst to best, calculate the z score of the person you're interested in, multiply by the SD (15) and add the mean.
It is also the case that for populations over 80 billion, you can have negative IQ scores, using the same logic that was used for a person with an IQ of >=200.
It means that the mean and standard distribution have been calibrated to a population, but that the population kurtosis is significantly non-normal
Incorrect. It's also possible if human intelligence isn't normally distributed.
Only if intelligence of the human population is normally distributed.
No you don't. You have invented this unnecessary step.
No, because the "person" and the z score have no link.
If a rock has zero intelligence, how can something score lower? Negative intelligence is impossible.
Look, I'm saying the same thing that I also found on Wikipedia. You just put the scores in order and then you fit them to a normal curve. This is what it means to scale them ordinally and then fit this to a normal distribution.
Its clear that we aren't going to agree on any of this, so I'm going to stop replying.
Additonally, you seem to incorrectly think that an IQ of 0 would mean zero intelligence when I have explained exactly what an IQ of zero would mean.
It's pretty obvious that a rock can't have an IQ of anything but zero.