Wednesday 26 September 2012

On the inequality of SATs


The original thread of discusison.

"Conversely, there are feminists who don’t understand distributions either,"

Never attribute to stupidity what which can be attributed to malice when the people in question have an obligation to not be stupid.

"Britain’s GCSE has girls outperforming boys across the board. "

Boys have been outperforming girls in maths since the coursework was taken off, viz. 2010. From wikipedia:

'Pirie also observes that the GCSE focus on coursework has disadvantaged boys reversing the gender gap in attainment, to the degree where in all subjects girls outperform boys, including traditionally male subjects such as sciences and physical education.'

Also interesting, the grade inflation in the said exams (and the A-levels) and the recent Michael Gove's crusade against the dumbed-down GSCEs.

"Interestingly, girls flat out WHOMP boys in Chinese and English. Just sayin’."

How's that whomping in the top percentile? or even top ten percentile?

"Interestingly, the trend reverses for the relatively newer “Writing” portion of the test."

It was included precisely because of that, making the test less sexist. PSAT's writing portion has earned many more girls scholarships than doubling the weight of the verbal scores.(which funnily did nothing for the top percentile)

"However, give the large numbers, there are more than enough girls and boys to fill the positions requiring these skills."

That sheen is taken off when you consider:

"1990 In this year, ten students out of 1.2 million test takers (roughly one in 120,000 students) get perfect scores of 1600 on the SAT. "

"In 1994, 25 students got perfect scores out of about 1.25 million (about 1 in 50,000 students). The first recentered SAT in April has 137 perfect scores out of about 200,000 test takers (about 1 in 1,400 students). "

"Out of 1.38 million seniors taking the SAT, 238 (roughly 1 in 5,000 students) receive a perfect score of 2400. In 2004, approximately the same number of seniors took the SAT, and 939 (about 1 in 1,500 students) received a perfect score of 1600. " 

and look at this graph of SAT scores over time.



The people who are in the no-maths-gap-and-due-to-sexism camp, use the fact that the boy-girl ratio used to be 13 to 1 in SAT for >700 scores and is now around 2 to 1 as proof that it can be whittled down to nothing.

"Interestingly, one could also look at post-admissions data for anti-boy bias (if admitted boys have higher scores, then perhaps they faced a higher admissions hurdle). "

That would be so if the rejected boys didn't have higher scores than rejected girls.

2005-2006:
7608 males applied, 10% admitted: 758. Number DENIED: 6850
2832 females applied, 26% admitteD: 736. Number DENIED: 2096


It's interesting to view these figures in the context of Simpson's Paradox(look for the Berkeley gender bias case at wikipedia). And that the almost-puking lady in Larry Summers's brouhaha came from MIT and Christina Sommers has written something about her past work in bringing about gender equality at MIT. And that the Dean of Admissions mentioned in above link was recruited for gender-equality, got some awards and MIT folks had a party for accomplishing this feat.

"the tendency for boys to be more likely, when down to two choices, guess and move on rather than dither"

and girls can't be taught this? For all the great steps they have taken in girl education, all that schooling changes and hard work they put in while the boys twiddle their thumbs on video games, and they can't be taught how to beat some sexist multiple-choice test!

No comments:

Post a Comment