scoring SAT scores (again)

over at Volokh ...

SAT scores: Cathy Seipp properly criticizes an anti-SAT piece in the L.A. Times. I'm not an expert on the SAT, but to the best of my knowlege the Times op-ed has still more problems. Just for starters,

1. Considering GPAs instead of SATs risks favoring students who go to schools where there's a great deal of grade inflation.



2. Letting in the top 10% of each high school will let in students who are quite academically weak, but went to a high school where the other students are weaker.



3. The result is not only bad for the intellectual quality of the college, but is also no service to the weaker students who will now find themselves at the bottom of the class, competing against others who are much more prepared for college.



4. The SATs coupled with GPAs are actually quite good predictors of college performance, for all the criticism that the SAT has drawn, and to my knowledge are much better predictors than GPAs alone.



5. That "white students score 206 points higher on average than nonwhites" on the SAT isn't a sign of bias in the SAT -- I am told that it actually slightly overpredicts college performance by black and Hispanic students -- but rather a sign of the unfortunate underpreparedness of many black and Hispanic high school graduates. This is a problem that needs attention, but letting in underprepared students into demanding college programs is probably not a solution.