I’ve been over a big educational attainment GWAS, and one of the main problems with them seems to me to be that they make you think that the amount of schooling a human gets is somehow a function of their personal biochemistry.
If you really want to look at this, you need to model social effects like availability, quality, and affordability of education, the different mind shapes needed to do well in school for people who are oppressed to different degrees or in different ways, whether people have access to education modalities or techniques shaped to fit their mind, whether the kid is super tall and gets distracted from grad school by a promising career in professional basketball, whether or not their mental illnesses are given proper care, and so on. If you measure how many years of education are afforded to a random human you mostly get social factors.
If you’re looking at the same big EA GWAS that threw out all non-Europeans that I’m thinking of, they didn’t look at any of that. I don’t believe a sufficient model is common practice, because as noted in the thread there is effectively no applied branch of the field that would expose the insufficiency of the common models.
I’ve been over a big educational attainment GWAS, and one of the main problems with them seems to me to be that they make you think that the amount of schooling a human gets is somehow a function of their personal biochemistry.
If you really want to look at this, you need to model social effects like availability, quality, and affordability of education, the different mind shapes needed to do well in school for people who are oppressed to different degrees or in different ways, whether people have access to education modalities or techniques shaped to fit their mind, whether the kid is super tall and gets distracted from grad school by a promising career in professional basketball, whether or not their mental illnesses are given proper care, and so on. If you measure how many years of education are afforded to a random human you mostly get social factors.
If you’re looking at the same big EA GWAS that threw out all non-Europeans that I’m thinking of, they didn’t look at any of that. I don’t believe a sufficient model is common practice, because as noted in the thread there is effectively no applied branch of the field that would expose the insufficiency of the common models.