Conditional on intellectual ability alone, maybe I’m only 1 in 20 (or 1 in a 100 [I haven’t taken an official IQ test, but I’d bet at 4:1 odds that I was above the 95th percentile but below the 99.9th percentile]). But I have more useful traits than just my raw intelligence. Considering other factors like parental socioeconomic status, epistemics, general and technical knowledge, rationality, etc. I think the class of people who are pareto better than me is probably something like 1 in 100,000 to 1 in 10 million.
That seems unlikely to me. Given that IQ is heritable, it tends to correlate with parental socioeconomic status. The others correlate as well. A 1 in 20 or even 1 in 100 IQ would put you below the average on LessWrong when we still had a census that was gathering IQ data.
(Maybe I was just being too conservative. Those are lower bounds I was confident betting on.)
I scored higher than the LW/SSC median on the IQ test that was linked (a Mensa Denmark one IIRC) in one of the surveys (SSC I think, I joined LW in 2017).
As I understand it, the online IQ test correlated with actual scores.
I may have been underestimating:
I had the highest scores in my IGCSEs of all my classmates
I found myself exceptional even in my undergraduate CS class (a population of over a hundred), etc.
I don’t honestly believe that I’m below the LW median. At least, I won’t willingly take such a bet.
(I haven’t taken an official test yet, boasting about IQ is gauche and doubly so for imaginary IQ, I’m conservative when betting, etc.)
FWIW, I originally wrote this for the EA Forum (whose norms around this I’m less familiar with, so I defaulted to conservativeness in bragging about intelligence) and didn’t edit it for LessWrong. Maybe I’d have been less hesitant to boast if I was directing it at a LW audience.
To me claiming to be 1 in 100,000 to 1 in 10 million overall seems like a pretty boasty claim with a higher boastiness than claiming 140 IQ or something on that order.
Actually, I think your comment about strong positive correlation between those traits is something that I may have insufficiently appreciated enough when writing the post.
That seems unlikely to me. Given that IQ is heritable, it tends to correlate with parental socioeconomic status. The others correlate as well. A 1 in 20 or even 1 in 100 IQ would put you below the average on LessWrong when we still had a census that was gathering IQ data.
(Maybe I was just being too conservative. Those are lower bounds I was confident betting on.)
I scored higher than the LW/SSC median on the IQ test that was linked (a Mensa Denmark one IIRC) in one of the surveys (SSC I think, I joined LW in 2017).
As I understand it, the online IQ test correlated with actual scores.
I may have been underestimating:
I had the highest scores in my IGCSEs of all my classmates
I found myself exceptional even in my undergraduate CS class (a population of over a hundred), etc.
I don’t honestly believe that I’m below the LW median. At least, I won’t willingly take such a bet.
(I haven’t taken an official test yet, boasting about IQ is gauche and doubly so for imaginary IQ, I’m conservative when betting, etc.)
FWIW, I originally wrote this for the EA Forum (whose norms around this I’m less familiar with, so I defaulted to conservativeness in bragging about intelligence) and didn’t edit it for LessWrong. Maybe I’d have been less hesitant to boast if I was directing it at a LW audience.
To me claiming to be 1 in 100,000 to 1 in 10 million overall seems like a pretty boasty claim with a higher boastiness than claiming 140 IQ or something on that order.
Actually, I think your comment about strong positive correlation between those traits is something that I may have insufficiently appreciated enough when writing the post.
I’ll just strip that segment out I think.