I’ve argued several times — for example, in my review of Charles Murray’s Real Education — that it makes little sense to judge the American educational system by comparing its students’ scores with the scores of students in other countries. America’s educational problems primarily have to do with underperforming minority groups, specifically blacks and Hispanics; one can say this is shameful, but it won’t solve the problem to look to countries that don’t even have underperforming minorities for guidance. When score comparisons take race into account, the U.S. does basically fine. Various sources support this analysis.
A new report from Eric Hanushek and two coauthors contradicts this, however. The report finds that even relative to American whites, a higher percentage of students in other developed countries earn very high math scores. Of course, this says nothing about average scores, but it’s an important point nonetheless.
The only potential problem I see is that this still isn’t an apples-to-apples comparison — the authors didn’t compare our white kids to their white kids, but our white kids to all their kids. This means their underperforming minorities are included while ours are excluded, but it also means their overperforming minorities are included while ours are excluded (“minorities” being used in the American sense; Asians are a majority in Asia, obviously). This is especially noticeable when you look at the very top countries: Taiwan, Hong Kong, and Korea. American Asians score higher than American whites on math, so it can hardly be blamed on American education if Asian Asians score higher than American whites, too.
What’s especially interesting is that American whites’ scores are quite comparable to those of students in the United Kingdom — the UK being the 92 percent white country from which America drew much of its population and even more of its culture.
Also interesting are the geographic disparities in white scores — whites in some American states score better than whites in others, and there’s a similar variance among European countries. I would like to see a truly apples-to-apples comparison, though the data would be rather difficult to collect. I’d like to see the students broken down by their countries (not merely continents) of ancestry, and their scores compared with those of students from their ancestral homelands. Do Americans of German ancestry score better than modern Germans? What about Finnish Americans?