Many of you don’t sufficiently appreciate the impact of an average IQ difference of 15 points between two populations.
This is common—many people don’t really “get” the effects of normal distributions at the tails—and Yale is a tail (har har, it rhymes). Tail effects are why the average man may only run a bit faster than the average woman, but there are thousands of men who can run a 100m dash faster than the women’s world record of 10.49 seconds.
IQ, by construction, is normally distributed—with a mean of 100, and standard deviation of 15. Cognitive ability is not perfectly normally distributed, but the normal distribution generally works quite well to model it (same deal with traits like height). Among large population groups, IQ appears to have a standard deviation of roughly 15, as well.
If we go with a standard deviation of 15 for each group, and if the mean US black IQ is 85, US gentile whites 100, and US Ashkenazi Jews 115—only .003% of blacks will have an IQ over 145, but .135% of gentile whites will, and 2.28% of Ashkenazi Jews—since for blacks, a 145 IQ would be 4 standard deviations above average. For whites, it would be 3 standard deviations above average, and for Ashkenazi Jews, 2 standard deviations above average.
On a per capita basis, there would be about 42.6 whites to every black with an IQ over 145, and 16.9 Ashkenazi Jews to every gentile White.
Then let’s consider the respective U.S. population size estimates of 38.9, 197.9 (maybe this could be 197.9 – 5.5 = 192.4, but shouldn’t make a big difference), and 5.5 million for blacks, gentile whites, and Ashkenazi’s, respectively (each from Wikipedia, I got 5.5 by averaging the 5-6 million range they gave).
Out of Americans who are black, gentile, or Ashkenazi Jews with an IQ over 145 (so not including Asians, etc.), blacks would only be 0.31% of said over-145 IQ group, whites 67.89%, and Ashkenazi Jews 31.80%. Almost no blacks, and about 2/3 gentile whites and 1/3 Ashkenazi Jews.
Now we can see why that in the absence of affirmative action, blacks would have very small percentages in any arena that selects for high IQ, and why Ashkenazi Jews would have a large presence—like at Yale.
However, if Ashkenazi’s really are 27% of the Yale population (as posted by someone) and whites overall are 62% (72% from the Yale site divided by the sum of all the percentages, since as noted they went over 100% because of Yale double-counting), this would mean a 35% gentile white population.
This seems like there’s more Ashkenazi’s Jews at Yale than you would expect based upon IQ figures, especially since an “IQ cutoff” to model Yale would probably be a bit lower than 145.
If we went with a 145 cut-off for Yale and we accept the numbers of 27% for Yale Ashkenazi’s and the figures on Yale’s site, A 27% to 35% ratio of .77 is higher than the 31.80% to 67.89% ratio of .468. This may suggest that Yale favors Ashkenazi Jews to gentile whites—the ratio of Ashkenazi Jews to gentile whites is .77/.468 = 1.65 times higher than we expected. The 1.65 number is conservative, since the 145 IQ cut-off for Yale is probably a bit too high. However, for context, the ratio of blacks to gentile whites at Yale is apparently 8.5%/35% = 0.24, whereas we would expect 0.31%/67.89% = 0.0046 based on the above IQ calculations. This means a ratio of blacks to gentile whites .24/.0046 = 52.56 times what we expected. Weirdly enough, if gentile whites really indeed only are 35% of Yale’s population, this would mean blacks might be over-represented vis a vis whites as 8.5%/35% = .242 > 38.9/197.9 = .1965 even if we pretended the blacks and whites have equivalent IQs—a .1965 ratio is what we would expect if we just randomly selected people with no regard to IQ. This would be pretty egregious.
Repeating the exercise with IQ over 160 leads to per capita ratios of 110.5 gentile whites to every black, 42.6 Ashkenazi Jews to gentile whites, and factoring in population numbers we have .08% blacks, 45.74% gentile whites, 54.18% Ashkenazi Jews—Ashkenazi’s would now outnumber gentile whites in the above 160 range.
Now we can see why it seems like every other historically eminent scientist is Ashkenazi Jewish (e.g. Einstein, Feynman, Von Neumann).
To the extent that low-IQ is associated with violent crime, repeating the exercise yet again for low-IQ would illuminate why violent crime rates by race of perpetrator are the way they are.
The above calculations are not meant to be law (for example, the US population figure are not age-adjusted, we can quibble about the average IQ estimates, cognitive ability is not perfectly normal, colleges select on things like conscientiousness in addition to IQ, etc.), but they help illustrate the general principle—the impact of differing mean IQs between population groups at the tails.