daver@hp-pcd.UUCP (05/19/84)
IQ tests have historically proved excellent predictors of success in the school system. IQ tests have also been strongly culturally biased (the last IQ test I took had vocabulary questions!). The problem is that success in the school system has been strongly culturally biased, and battles against the IQ tests have obscured the real problems with the school system. Dave Rabinowitz hplabs!hp-pcd!daver
csc@watmath.UUCP (Computer Sci Club) (05/20/84)
One often hears the statements "Tests can only measure the ability to take
tests" and "Tests can only predict the ability to take tests". The first
is tautological (in terms of the TQ analogy, A test of Tallness can only
measure a persons tallness) the second is simply false. Some tests are
in fact fairly good predictors.
Take the much abused I.Q. It is not a very good predictor of individual
preformance but it is not a bad predictor of group preformance. Example:
Take a large group of ten year olds. Task find a group of half of them who
will not get degrees in mathematics. You chose by random. I choose
the half with I.Q. less than 100. Guess who's prediction is better?
(Naturally neither prediction will be perfect, though mine will be very
good. The only perfect test of whether a person will get a degree in
mathematics is to wait around and see. Imperfect tests are used because
we have no other practical choice. In interpreting results it is always
necessary to remember that our tests are imperfect.)
Tests are only useful if they are correlated to some property we are
interested in. (I don't care how well you can write I.Q. tests, I do
care whether you will do well in university). If the correlation is
strong they become useful predictors. If the correlation is VERY (and
I mean VERY) strong they become useful predictors of individual preformance.
In most cases the correlation is not that high. This does not
make the test useless or uninteresting, it just means they are not
very good for predicting individual preformance.
William Hughes