I recently finished a long exchange, one originally on the biological nature of race, with the zealous philosopher Jonathan Kaplan. It ended, on his end, with him accusing me of being a "white supremacist" and exclaiming:

"Because one place we "disagree" is with respect to your claim that
"African and Amerindian peoples are more racially discriminated for
than against in the U.S." That claim is, to be blunt, completely
insane racist bullshit. If you really believe that, arguing with you
is completely pointless; you are so out of touch with the actual state
of the world that I am unable to trust anything that you say, nor can
I trust that your goals are at all reasonable."

Apparently, my argument that African and Hispanic Americans were more discriminated for than against was simply too unpalatable for him. I imagine that he didn't care for my subsequent reply, either. Anyways, the key issue, which we didn't get around to discussing in detail, was whether the outcomes differences could be more than explained by population level effects of IQ. My positions, following Robert Gordon's, was that were one takes into account, e.g., regress out the effect of, the IQs of parents, siblings, friends, neighbors, teachers, mayors, etc. in addition to that of individuals, the direction of group differences would reverse. Sociologists typically don't see this because at most they merely control of individual differences in IQ. And when they do control for some group measures they end up conceptualizing the differences in terms of "segregation".

The population level effect of IQ explains why IQ differences are so much more potent on the group (e.g., subcultural, ethnic, national) level than on the individual one. Simply, it's just much better to be low IQ (e.g. 85) in a moderate IQ extended network (family/neighborhood/region) (e.g., 100) than to be the same in a low IQ one.

Anyways, I was wondering if anyone was aware of a publicly accessible survey which contained family+ ability scores, which would allow me to show this. If I can't find anything better I will look at the Fragile Families Survey, which at least has parental+sibling IQ scores.
http://www.fragilefamilies.princeton.edu/ Alternatively, if someone knows of research on the matter already done, I would be much obliged if they pointed me to it.

Thanks.