A new Brookings Institution study has found that self-esteem is correlated with good school performance within countries, but inversely related to school performance between countries. That is, for example, Japanese kids have lower self-esteem and higher scores than American kids do, but within America or within Japan, the kids with the highest scores also have the highest self-esteem.

Julian Sanchez of Reason has a good analysis of the data here. He points out that Asian countries in general have high scores but cultures that look down on boastfulness, and that self-esteem is created by comparing yourself to your peers, not to people in foreign countries. So you’d never expect inter- and intra-national comparisons to yield the same results.

I’m not sold on his statistical argument though. He has us “imagine” two countries with two different distributions of talent, without giving any reason to believe these countries are anything like what we see in the real world.

The only thing I have to add to the discussion is some conservative pundits’ point that, within America but between ethnic groups, self-esteem and performance are inversely correlated. That is, black students — the most frequent target of “self-esteem initiatives” — actually have higher self-esteem than the white students who outscore them academically. Some then argue that these initiatives are actually counter-productive.

The most logical conclusions, consistent with all this evidence, are that (A) self-esteem is tied to one’s cultural prescriptions regarding boastfulness, and that (B) within those cultural prescriptions, self-esteem is tied to one’s abilities, either as a cause, an effect or both. Culture explains why American blacks have high and American Asians low self-esteem, and why inter-country comparisons bring out an inverse correlation, while actual performance explains the variations within countries and ethnic groups.

Robert VerBruggen blogs at http://robertsrationale.blogspot.com.