fair, at least that would suggest that fewer than half of all people are less smart than average though
Yeah, thatâs not what Iâm talking about.
To your point, i think iâve read our ability for different kinds of thinking peaks at different ages.
Interestingly, chess has been measured, using chess AI to establish a norm for measuring the optimal move.
Itâs 35-45. Thatâs for grand masters. Most of us can get better at any age.
Iâd guess itâs âsymmetricâ by definition. Like the final scores have to be arbitrary, donât they?
Unless we have some external definition of IQ? Number of teraflops required to fully emulate the human?
1 Like
They could be arbitrary but non-symmetric, no? If itâs forced symmetric then 1 IQ point wouldnât necessarily be equal at different points along the distribution (I think).
1 Like
Sure, It could arbitrarily be non symmetric⌠But (as far as I know) 1 point of IQ is always arbitrary, so it can be arbitrarily forced into symmetry.
That is, I think thereâs no meaning to the word âequalâ in your sentence.
Surely it measures some quantifiable amount of IQ (hence the quantification)?
Imagine a score in a math test.
The score measures definite knowledge.
But whether the score is symmetric is in a sense arbitrary. This is because the mapping of the knowledge to any kind of score is conventional.
One way to set the convention is to measure it in standard deviations from the mean. That has the potential to be asymmetric, if the scores have a heavy tail.
But if you calibrate by percentile then it will have to be symmetric.
Itâs not clear to me that either is more correct.
There is an old fallacy that a gaussian distribution implies there is some real quantity underneath you are measuring. This follows from inverting an error model from astronomy and physics: a central true value subject to lots of little perturbations will give a gaussian distribution around the new plus some bias. But the reverse doesnât hold; a gaussian doesnât mean there is some true value underneath.
Under this fallacy, measuring in standard deviations, and finding a gaussian distribution, tells you that you have measured some t
âtrueâ latent quantity called âmath knowledge.â And finding asymmetry then must tell you something too.
But as I said thatâs a fallacy.
Ok pretty sure my IQ is in the 90âs. I clicked on not one but TWO âfree IQ testsâ and completed them before they asked for money. I gave one an email address I use mostly for junk. It said my IQ was 148 and for $$ I could get additional info. I recognize that the very fact that I clicked means my IQ is at least 50 points less than that. But since I didnât give them $ I think I might be 90 or above.
3 Likes
IQ = Intelligent Quotient; in modern parlance, itâd be referred to as an Index.
So an IQ of 100 in 1965 isnât going mean the same as an IQ of 100 in 1995. That is, these two scores are not comparable in any appropriate fashion.
Not giving them money is definitely a good sign for your IQ score
1 Like
Agreed, this is a good point. Iâd read a bit about this on the IQ Wikipedia page, there are misconceptions that peopleâs IQ degrades with age but apparently a better explanation has simply been that successive cohorts of people have tended to have higher IQs, so you just have a lower relative IQ.
get yer smartz outta this here IQ thread
From my direct experience, intelligence on the lower end is not easily measurable⌠as in, they have trouble understanding the test items and canât answer questions at all.
Once you get to that sort of tail, youâre getting into a simple checklist territory, trying to figure out what sort of skills they do have â can they talk? are they able to dress themselves? Can they follow certain types of instructions?
A number is not exactly helpful to people - you have to get specific about the limitations.
6 Likes
I finished it when I took it for my first employer. I never asked what my score was though. I did prep for it the day before to make sure I finished.
Are you⌠talking about actuaries here?
People were talking about the distribution/tails.
I know people are generally talking about the right tails, but Iâve got a lot of experience with the other end, too. I have a lot of special education teachers in my own family, and then thereâs my own son. He scores 3rd percentile on whatever tests⌠I donât mind them having him take the state tests (somebodyâs got to be the 3rd percentile), but I donât much see the point.
3 Likes
INTP
And Iâm at least 100. Maybe higher.
1 Like
3 Likes
But getting to the tail issue, norming the IQ scores, etc.
Thinking about what the tests were created for, and that they keep having to get re-normed due to the Flynn Effect, most tests are probably worthless for the tails anyway (top/bottom 2%), either side.
I mentioned what itâs like on the low end â on that side, youâre going to be needing a high aide-to-student ratio as it is, and youâre going to be approaching very individualized training. These students have IEPs, and many will never be able to live independently.
On the other end, they may be able to self-teach many things (or be able to get very far in this internet age w/ online courses, etc., if theyâre self-paced), and there have been special programs created for these types of students (like CTY and TIP). But trying to pin down exactly where they are in that 2% is not all that important.
The IQ tests work best for the 96% in the middle, so talking about the tails is somewhat beside the point.
1 Like