Arnold Kling  

Manski and Caplan

PRINT
Virtual Meetup... Thomas Sargent on Government D...

I just got around to reading Charles Manski on "Genes, Eyeglasses, and Social Policy." Manski is dismissive of heritability studies, and I am curious how Bryan would react, given the importance he places on heritability studies for his argument that parents should relax. Manski quotes Arthur Goldberger's sarcastic comment,


Now suppose that the entire population lacks eyeglasses. Then the heritability of effective quality of sight is one. What does this imply about the usefulness of distributing eyeglasses as a treatment for nearsightedness? Nothing, of course.

What I believe Manski wants to argue is that high heritability of IQ does not mean that IQ is untreatable, just as high heritability of nearsightedness in a population without eyeglasses does not mean that nearsightedness is untreatable.

What I would counter is that high heritability of a trait may not tell us the effectiveness of treatments that have yet to be discovered. However, it does tell us that the existing treatments in a population make little difference.

As an aside, the eyeglass example is peculiar in that people seek finite absolute status rather than relative status or maximum status. The ability of eygelasses to equalize vision across the population might disappear if parents of children with good vision sought "gifted and talented" eyeglasses to enhance their children's eyesight even further.


Comments and Sharing





COMMENTS (3 to date)
rapscallion writes:

It's yet another PC anti-IQ piece that completely ignores mounds of evidence while demeaning IQ researchers with dismissive, condescending language that would never be used against even ridiculously fanciful ideas that were PC-compatible. For example, "Given that it was widely recognized more than 30 years ago that heritability research is irrelevant to policy, I find it both remarkable and disheartening..."

Basically, Manski's shtick is to write as though IQ researchers have widely argued: i) IQ is heritable. ii) Nothing heritable can be treated. iii) Therefore, IQ cannot be treated.

No prominent IQ researcher has ever offered no evidence except heritability to argue that social policy can't do much to raise IQ. He writes as though decades of research haven't failed to find effective treatments. He beats a straw man to death.

Jason Collins writes:

Manski does not take the argument any further than Goldberger and manages to avoid 30 years of responses to Golderger's critique. I posted a response to Manski's piece, but recommend the article by Beauchamp and colleagues from the same issue of the JEP.

Michael Bishop writes:
What I would counter is that high heritability of a trait may not tell us the effectiveness of treatments that have yet to be discovered. However, it does tell us that the existing treatments in a population make little difference.

I can interpret this in a way which is correct, but I'm not sure exactly what you meant and I definitely think others will interpret it incorrectly. I would say that high heritability does not mean existing treatments make little difference, but rather that VERY high heritability means that VARIATION in treatments explain little of the VARIATION in outcomes. Heritability is far from irrelevant, but it needs to combined with other information to translate it into policy implications.

Comments for this entry have been closed
Return to top