[talk.politics.theory] Bias on IQ tests

sethg@athena.mit.edu (Seth A. Gordon) (03/22/88)

In article <3933@bloom-beacon.MIT.EDU> 
lazarus@athena.mit.edu (Michael Friedman) writes:

>It is often claimed that blacks do worse than whites on tests like the
>SAT's and Civil Service tests, etc. because the tests are biased. Does
>anyone out there have a list of supposedly biased questions?

I suspect it's more class than race.  Middle- and upper-class people are
simply more exposed, in their culture, to the sort of things the tests
ask about than lower-class people are.  Since the average income level
of blacks is lower than that for whites, blacks tend to score lower.  I
don't know what the stats are for black vs. white when you correct for
economic level.

I heard an interesting anecdote from my psych. class on men and women.
When a draft of the Stanford-Binet (I think) IQ test was tested, it was
discovered that women tended to score *higher* than men on it.

The designers of the test *removed* some of the questions that women
tended to score better on, so that men and women would average the same
score.

This speaks volumes to me on the relationship between "intelligence,"
whatever that is, and intelligence tests.

-- 
sethg%athena.MIT.EDU@mit-eddie.UUCP -- CONVERT me, CONTRA lovers! --
sethg%athena.MIT.EDU@mitvma.BITNET| talk.politics.latin-america: YES 40 / NO 3
sethg@athena.MIT.EDU -------------| I need **63** more YES votes by March 31.