Hacker News Reader

Comment by barry-cotter | Hacker News Reader

Griggs vs. Duke Power Company

http://en.wikipedia.org/wiki/Griggs_v._Duke_Power_Co.#Judgme...

Since the SAT[1] is a glorified IQ test[2] the rankings from high to low by population group go Ashkenazi Jews >> East Asians >> Caucasian >> Blacks.

The Disparate Impact legal doctrine[3] means that unless you can show business necessity any test that reliably discriminates between protected classes[4] is automatically illegal.

The strength of this argument is weakened by the fact that TTBOMK no equivalent law exists in the rest of the Anglosphere.

[1]And the LSAT, MSAT, GRE, GMAT etc.

[2]This isn't completely accurate for those tests that measure concrete knowledge like the MCAT, but the correlation is still pretty high

[3]http://en.wikipedia.org/wiki/Disparate_Impact

[4]http://en.wikipedia.org/wiki/Protected_class

View on HN

Thanks for the detailed reply. It's funny that you can't use a test to select people for a job, while you can use the same test to select people for college (though perhaps I'm extrapolating, since the wikipedia article only mentions that you can't use the test for internal transfers and promotions.)

Since the SAT[1] is a glorified IQ test[2]

The SAT is a wonderful test for memorization. Also how much time you spent studying, almost as if its a... scholastic test.

It it's relationship to IQ is extremely vague.

http://pss.sagepub.com/content/15/6/373.abstract

This research established the relationship between SAT and g, as well as the appropriateness of the SAT as a measure of g, and examined the SAT as a premorbid measure of intelligence.

In Study 1, we used the National Longitudinal Survey of Youth 1979. Measures of g were extracted from the Armed Services Vocational Aptitude Battery and correlated with SAT scores of 917 participants. The resulting correlation was .82 (.86 corrected for nonlinearity).

Study 2 investigated the correlation between revised and recentered SAT scores and scores on the Raven's Advanced Progressive Matrices among 104 undergraduates. The resulting correlation was .483 (.72 corrected for restricted range). These studies indicate that the SAT is mainly a test of g.

We provide equations for converting SAT scores to estimated IQs; such conversion could be useful for estimating premorbid IQ or conducting individual difference research with college students.

I still recall a very brief and elegant point my pedagogy professor made by giving all of his students an IQ test.

The test just happened to test on material none of had ever studied, for example how to clean offal. This was a real test which people with the correct experience had scored highly on. Us college students on the other hand, were deemed barely functional by the same test.

The IQ test is not contrived but it does rely on learned knowledge to be able to test. So the IQ test itself is not a very good test of intelligence, especially in the case where the tool (lets say language) it's using to test IQ is lacking in say non-native speakers.

But to get back to the SAT test, my East Asian immigrant high-school friends with decent English - bad SAT scores. Two years later with greatly improved English, same people - great SAT scores. And I am talking about all of the SAT, including the math part.

Reply on news.ycombinator.com