Is this the post you’re looking to cite? From what I’ve been able to tell looking at that post and some related sources physicists seem to be about 2 standard deviations above the norm on average, with an uncertainty in that average measurement of a couple IQ points. This makes a physicist something like 1 in 50, rather than 1 in 100,000. Unfortunately it’s not really clear what the standard deviation is for those numbers. An average, by itself tells you very little about a distribution. If we assume that the standard deviation of IQ scores of physics Ph.D.s is close to that of the rest of the population (this may be a bad assumption) then we expect 10% of physics doctorate holders to have an IQ of 110 or less, which would put them outside of the top quartile of the IQ distribution. You could play around with the parameters for the distribution here and come up with a bunch of different results, and I’m not totally sure which ones are meaningful.
I may be wrong. I may be framing the problem wrong. I may not have a good picture of just how bad merely average quantitative skills are. I do work with a lot of people who have Ph.Ds in physics and while some of them certainly have remarkable quantitative skills, others really don’t. I know people who have gotten doctorates in physics who didn’t score particularly well on the GRE. This is especially true in experimental physics (in physics the standard joke is that as a theorist you don’t have to study anything that exists in the real world, and as an experimentalist you don’t have to get the math right). Physicists are pretty smart people, but they’re not all 1 in 100k and that’s certainly not a requirement for being a physicist. Also, I think the level of quantitative ability necessary for physics research is often below the level needed to muddle through the coursework.
Is this the post you’re looking to cite? From what I’ve been able to tell looking at that post and some related sources physicists seem to be about 2 standard deviations above the norm on average, with an uncertainty in that average measurement of a couple IQ points. This makes a physicist something like 1 in 50, rather than 1 in 100,000. Unfortunately it’s not really clear what the standard deviation is for those numbers. An average, by itself tells you very little about a distribution. If we assume that the standard deviation of IQ scores of physics Ph.D.s is close to that of the rest of the population (this may be a bad assumption) then we expect 10% of physics doctorate holders to have an IQ of 110 or less, which would put them outside of the top quartile of the IQ distribution. You could play around with the parameters for the distribution here and come up with a bunch of different results, and I’m not totally sure which ones are meaningful.
I may be wrong. I may be framing the problem wrong. I may not have a good picture of just how bad merely average quantitative skills are. I do work with a lot of people who have Ph.Ds in physics and while some of them certainly have remarkable quantitative skills, others really don’t. I know people who have gotten doctorates in physics who didn’t score particularly well on the GRE. This is especially true in experimental physics (in physics the standard joke is that as a theorist you don’t have to study anything that exists in the real world, and as an experimentalist you don’t have to get the math right). Physicists are pretty smart people, but they’re not all 1 in 100k and that’s certainly not a requirement for being a physicist. Also, I think the level of quantitative ability necessary for physics research is often below the level needed to muddle through the coursework.