/sci/ - Science & Math

>>15923980
>While the individual effect sizes of any individual allele are, of course, small, still hundreds to thousands have been identified as functional variants and they do show differences in distributions between populations.
And the part people tend to miss is calculating what that difference actually amounts to and what proportion they're present in. Once you get to something like 50% within-population commonality you've extremely few variants, and if you're talking further approaching 80-100% it's a bit of a joke.

So what you effectively have to do is some statistics and figure out based on average effect sizes for variants what the frequency distributions would have to be. Pretty sure it's nothing like what is observed given how small the effect sizes are and proportionately how rare cognitively relevant alleles are. However that still doesn't help answer the more complicated question about systemic environmental differences because that alone does not tell you the reaction norm effect those variants have across environments. It is a lot more complicated than just "line say this therefore infer that".
>Twin studies demonstrate that a large portion of the difference in intelligence between individuals is genetic.
Yes? I pointed that out? A lot of things show that. Again the core problem is exactly that misunderstanding common among the public but well known for relevant academics. "Is genetic" does not mean "the same across environments" by default, especially if there are systemic differences. Though you can make some effort in estimating reaction norm in a sample and some people have but it's not easy.

Maybe a simple illustrative image would help so I grabbed the one off wikipedia. "Is genetic" or "highly heritable" does not by mere presence tell you anything about reaction norm, and estimating that requires a lot more work and often way more than anybody bothers even attempting.