Note that I’m not defending the original claim you responded to, because I don’t think it’s using great data. Test scores would work great only if almost every

Note that I’m not defending the original claim you responded to, because I don’t think it’s using great data. Test scores would work great only if almost everybody reached the final year of high school and took the test, or if it came with info about what proportion of the students took it, but it would only say anything about the region or country taking it, not directly about others. (What it could also do is help refute the applicability of historical third world IQ measurements in general, too. But only if you actually identify the exam and location by name.)