Check out page 10: http://www.amazon.com/gp/reader/0815736398/ref=sib_dp_pt#rea...

Unfortunately I don't have my copy on me so I can't look up the footnote in the back. IIRC it was the SAT, but I'm not entirely positive.

If you want to see something really telling about standardized testing though, check out this Greg Palast article on how the NCLB discriminates against minorities:

http://www.phillyblog.com/philly/nation/13968-no-child-left-...

That does not answer the question are they using a random selection of 13 year olds vs a random selection of 17 year olds or a self selecting group of 13 year olds that take the text vs a random selection of 17 year olds who take the test.

If the methodology was that severely flawed they obviously wouldn't use the study in the book. This isn't Oprah's book club, it's a report co-authored by multiple Ph.D.s that was commissioned by the Brookings Institution.

Aren't you judging the kind of thing that makes credentials meaningful by citing the credentials themselves? As high finglefarb of the church of deontoenails, I am fully able to condemn this as heretical. And as someone experienced with condemning people as heretics, I am obviously qualified to be the high finglefarb.

Good question: "That is an interesting finding, I'd be interesting in seeing the methodology behind the study."

Insulting question: I wonder if the authors made a certain specific error, when making said error implies the person is either ignorant, stupid, or intellectually dishonest.

I'm not not going to defend the statistic since I haven't seen the methodology. That said, I do think it's insulting to the authors to ask if they've made a very basic and specific error without first reading the methodology.

In any event, I'm sure you could email any of the authors and ask where the statistic came from if you are interested.