Stop the stereotype, not the tests
GREAT post by John over at Discriminations (from last week - I've been out of the loop for a while) on stereotype threat and the fallacious argument that creating different testing standards for minorities is a useful way to deal with the threat. Go read it now, and see whether you agree with John's conclusion that Claude Steele's policy recommendations would aggravate stereotype threat, rather than relieving it:
"Stereotype threat" means that blacks don't do well on standardized tests where there graders are aware of racial differences in performance on standardized tests. Thus it would seem to follow that race-blind admissions -- where the "graders" did not know the race of the applicants -- would be reasonable solution, if "stereotype threat" is the problem.
Steele does not recommend that. Quite the opposite: he actually recommends discounting the test results for blacks, thus re-inforcing the notion (or confirming the stereotype) that they do not do as well. In fact, he submitted expert testimony supporting the University of Michigan's argument that standards have to be lowered for blacks.
I love John's sarcastic conclusion:
Do any other groups besides blacks, Hispanics, and Native Americans -- hillbillies, crackers, rednecks come to mind (and I use those terms positively, not negatively) suffer from "stereotype threat"? If so, shouldn't they get a break, too? What of Asians? Do they have a "stereotype boost"? (They are expected to do well on tests, so they do.) If so, should they be penalized?
Oops, I almost forgot. They are penalized.
If you aren't already reading John every day, why on earth not? Don't believe those people who tell you to get off the internet and "get a life". Go read Discriminations instead.