Am I reading the linked example correctly, that Asian-Americans’ IQ keeps growing during the last decade, and everyone else’s IQ in USA keeps dropping?
I mean, if you assume that the differences is SAT scores between races reflect their differences in IQ, it seems reasonable to assume that the differences in SAT scores between now and ten years ago reflect the differences in IQ between now and ten years ago. (Either that, or SAT also reflects something else beyond IQ.)
First, look at the magnitudes. The difference between Asians and blacks is about 380 points in 2015. Compared to that number the declines (-6 to −28) are very minor.
Second, while I don’t have data at hand, I strongly suspect that the Asian population of SAT takers changed during the last decade. In particular, the upper class of China got wealthy enough and “international” enough to start sending their kids to US universities (which usually involves taking the SAT) and that’s besides increased immigration from China in general.
Third, SAT is a proxy for IQ and it’s not a stable test. It’s being tweaked and adjusted constantly. Among other things, SAT is normalized so that the score of 500 corresponds to about the 50th percentile of test-takers. If you have a influx of smarter-than-average kids, they will not only push their subgroup scores up, they will also push everyone else’s scores down.
The SAT scores from different years are somewhat comparable (because they are normalized), but not fully comparable because the tests from these different years are literally different. That’s not a problem for comparing the performance of subgroups in any given year, though.
Am I reading the linked example correctly, that Asian-Americans’ IQ keeps growing during the last decade, and everyone else’s IQ in USA keeps dropping?
I mean, if you assume that the differences is SAT scores between races reflect their differences in IQ, it seems reasonable to assume that the differences in SAT scores between now and ten years ago reflect the differences in IQ between now and ten years ago. (Either that, or SAT also reflects something else beyond IQ.)
Several things here.
First, look at the magnitudes. The difference between Asians and blacks is about 380 points in 2015. Compared to that number the declines (-6 to −28) are very minor.
Second, while I don’t have data at hand, I strongly suspect that the Asian population of SAT takers changed during the last decade. In particular, the upper class of China got wealthy enough and “international” enough to start sending their kids to US universities (which usually involves taking the SAT) and that’s besides increased immigration from China in general.
Third, SAT is a proxy for IQ and it’s not a stable test. It’s being tweaked and adjusted constantly. Among other things, SAT is normalized so that the score of 500 corresponds to about the 50th percentile of test-takers. If you have a influx of smarter-than-average kids, they will not only push their subgroup scores up, they will also push everyone else’s scores down.
The SAT scores from different years are somewhat comparable (because they are normalized), but not fully comparable because the tests from these different years are literally different. That’s not a problem for comparing the performance of subgroups in any given year, though.