Gary Rubinstein has produced a stunning analysis of New York City’s high school report cards, its so-called progress reports.
He asks: “Why does the ‘worst’ NYC high school have higher SAT scores than the ‘best’ one?”
This is what Gary found: the SAT scores of the city’s highest-rated high school are lower than those of its lowest-rated school.
Read that again.
Maybe you don’t think much of SAT scores. But then look at those report cards again, and you will see some very unimpressive high schools–by any measure–ranked far above the city’s top high schools.
What a fraud these report cards are.
Diane Ravitch's blog 
The link says it isn’t available, try later.
Did we crash the server already?! It’s not even 7 AM yet.
I fixed the link.
I encountered the same problem. Hmmmmm….. Even the most innocuous things take on a different interpretation when it involves the big ( or not so big in the case of Bloomberg), the powerful and the morally bankrupt being viewed in less than glamorous light.
I fixed the link to Gary Rubinstein’s blog.
You can go to his site directly if you cut and paste in your browser.
http://garyrubinstein.teachforus.org/
It is up now. I was interested in seeing what percentage of students took the SAT at each school, and Mr. Rubinstein does report it. The top rated school had 55.3% of the four year cohort take the SAT with average scores of 325, 325, 345, while the failing school had 36.5% of students take the SAT with average scores of 373, 362, 356.
As Mr Rubinstein says, the differing percentages taking the SAT at each school make a direct comparison difficult and it would be a good idea to try to control for other factors known to influence SAT scores like family income, but I agree that this finding suggests that these schools are doing an equally ineffective job preparing students to go to college. A large fraction of these students would be admitted to my University (we accept anyone who graduates in the top third of their high school class), but I doubt many could possibly succeed.
” I doubt many could possibly succeed… schools are doing an equally ineffective job preparing students to go to college.”
SAT scores have poor predictive validity for college success as indicated by years of research. Based on this replicated data, about 400 colleges and universities have dropped the SAT requirement for admissions. http://www.fairtest.org/satvalidity.html
It’s an unsubstantiated leap of logic to extend SAT scores as an indicator of preparation for college. Those scores indicate little to nothing of the attributes students require to succeed in college.
I do agree that differences of 200 or 300 points in math and verbal SAT scores do not necessarily mean very much for likelihood of success in college, but larger differences do seem to matter on average.
The 660 score would put a student in the lowest 5% of SAT scores. The highest 5% would require a combined score of about 1400. I think that the student with a 1400 has a much higher chance of making it to the second year of college and graduating in 6 years or less than the student with a 660 score.