If we run the regression y = a*x + b*z + u1 and z = c*x + u2, shouldn't the coefficient "a" pick up all the effect of x? That is, if IQ_child = a*IQ_parent + b*_Income_parent + other_uncorrelated_stuff, shouldn't "a" capture the full effect of IQ_parent?
B-W gap exists in all countries irrespective of language country speaks, religion, political type, economical type, climate, nutrition e.g. Claiming there's no B-W genetic gap is extraordinary claim, and it should be supported by evidence, not vice versa.
Probably should point out that the small correlations depend on the variation in admixture(when multiple SIRE groups are used, including those with significant variation in admixture(such as hispnics and multiracial individuals in the US), the correlation is significantly higher(SIRE usually doesn't have too large effects, so i'm not as worried about confounding across Sire groups),(0.41 in TCP/PNC sample(Fuerst et al. 2021), and 0.37 in the ABCD sample).
>In hindsight, this is not surprising because these are only moderately useful proxies for genetic ancestry, and because within a given admixed group, the Pearson correlation between genetic ancestry and intelligence will be very weak (e.g. r = 0.09 in Lasker 2019), and this is true even if the entirety of the gap is caused by genetics. This is simply due to the range restriction in ancestry (i.e., variance reduction) and the large within population genetic variation in intelligence.
Low ancestry-iq correlations are compatible with large between-group heritabilities because the gap in ancestry between groups is enormous, being on the order of multiple standard deviations.
Difference in European ancestry is about 1-0.187=.817. In deviations, 6.95 d. So ancestry correlation of .09 predicts a gap of 0.09*6.95=0.63 d in IQ. Which is about 60% of the gap in this sample (1.01 d).
I think hereditarians should focus on creating predictive models that have a real-world impact. There is no end to pilpul from the communist side here.
If you take a step back and stop seeing the forest for the trees, something wonderful has happened. Uninformed normies treading into these waters for the first of course will still make the bread and butter arguments on SES and such, but this is easy to knock down, and such people will expect defenses from experts that they can point to, and among the people in the thick of the actual debate, the only people making arguments for 0% between-group heritability are these insane communist types, and even these people have long abandoned the basic lines of arguments for more esoteric critiques. When you look around and these are the only people who are on your side, you have to infer that something has gone awry.
Yes because it’s the proportion of observed statistical variability in relation to the transmissibility of an aggregate variable. When variability exceeds potential..
If we run the regression y = a*x + b*z + u1 and z = c*x + u2, shouldn't the coefficient "a" pick up all the effect of x? That is, if IQ_child = a*IQ_parent + b*_Income_parent + other_uncorrelated_stuff, shouldn't "a" capture the full effect of IQ_parent?
B-W gap exists in all countries irrespective of language country speaks, religion, political type, economical type, climate, nutrition e.g. Claiming there's no B-W genetic gap is extraordinary claim, and it should be supported by evidence, not vice versa.
Probably should point out that the small correlations depend on the variation in admixture(when multiple SIRE groups are used, including those with significant variation in admixture(such as hispnics and multiracial individuals in the US), the correlation is significantly higher(SIRE usually doesn't have too large effects, so i'm not as worried about confounding across Sire groups),(0.41 in TCP/PNC sample(Fuerst et al. 2021), and 0.37 in the ABCD sample).
>In hindsight, this is not surprising because these are only moderately useful proxies for genetic ancestry, and because within a given admixed group, the Pearson correlation between genetic ancestry and intelligence will be very weak (e.g. r = 0.09 in Lasker 2019), and this is true even if the entirety of the gap is caused by genetics. This is simply due to the range restriction in ancestry (i.e., variance reduction) and the large within population genetic variation in intelligence.
Low ancestry-iq correlations are compatible with large between-group heritabilities because the gap in ancestry between groups is enormous, being on the order of multiple standard deviations.
PNC Lasker 2019 data:
Blacks mean/SD: 0.187, 0.117
Difference in European ancestry is about 1-0.187=.817. In deviations, 6.95 d. So ancestry correlation of .09 predicts a gap of 0.09*6.95=0.63 d in IQ. Which is about 60% of the gap in this sample (1.01 d).
It's all so tedious, Emil.
I think hereditarians should focus on creating predictive models that have a real-world impact. There is no end to pilpul from the communist side here.
If you take a step back and stop seeing the forest for the trees, something wonderful has happened. Uninformed normies treading into these waters for the first of course will still make the bread and butter arguments on SES and such, but this is easy to knock down, and such people will expect defenses from experts that they can point to, and among the people in the thick of the actual debate, the only people making arguments for 0% between-group heritability are these insane communist types, and even these people have long abandoned the basic lines of arguments for more esoteric critiques. When you look around and these are the only people who are on your side, you have to infer that something has gone awry.
> between group heritability is a weird metric. It can be larger than 100%
You're the only other person I've ever encountered who understood this.
Yes because it’s the proportion of observed statistical variability in relation to the transmissibility of an aggregate variable. When variability exceeds potential..