Not only that, but you ignore the fact that you said, A much better indicator would be raw SAT rates or some other indicator
Raw SAT qualifies them in the first place.
They were selected from 15,000 finalists and 1.3 million high school students who entered the program by taking the 2002 Preliminary SAT/National Merit Scholarship Qualifying Test as juniors.
All finalists compete for one of the National Merit $2,500 Scholarships. The number of winners in each state is based on its percentage of the national total of graduating seniors.
A selection committee that included college admissions officers and high school counselors reviewed the information submitted by finalists and their schools, evaluating each finalist's academic record, scores from two standardized tests, leadership, essay and a written recommendation.
I have not "overlooked" that at all. Of course they must. But the point is that a school can have a ton of high-scoring SAT students, and still end up with fewer National Merit Scholars than a school which has fewer high-scoring SAT students, if the poorer school more aggressively works to get their qualifying students *applied* than the better school.
Thus my original point: The number of NMS scholars is a poor indicator of educational quality, since there is a factor involved (application rates) which can significantly affect the results for a given school. Instead, let's just compare SAT scores *directly*, and not number of NMS scholar finalists as you have attempted to do.
They qualify based on the test, and then further consideration is based n a combination of tests, essay, academics, etc.
...none of which will results in an NMS finalist if the student or school drops the ball and doesn't bother to ensure that every qualifying student actually SENDS IN AN APPLICATION.
Not only that, but you ignore the fact that you said, A much better indicator would be raw SAT rates or some other indicator
Where in the hell do you hallucinate that I "ignored" what I wrote?
Raw SAT qualifies them in the first place.
Yes, I know. But that still doesn't change the points I've raised. If you're still unclear on the point I'm making as to why NMS numbers are not necessarily a good indication of school quality due to the statistically biasing effect of self-selection, try reading it again until the little light goes on.