From: Dan Fabulich (daniel.fabulich@yale.edu)
Date: Thu Oct 07 1999 - 12:05:04 MDT
'What is your name?' 'Lee Daniel Crocker.' 'IT DOESN'T MATTER WHAT YOUR
NAME IS!!!':
> Just to clarify, I have never, and would never, argue in favor of
> "standardized" exams.
Perhaps you aren't committed to exams with a questions packet and a
fill-in-the-bubbles answer sheet, requiring a no.2 pencil, but any GIVEN
organization is very likely to use "standardized" exams in some sense or
other. Whether they're fill-in-the-bubble exams or whatever, they still
suffer from the same problem I was discussing earlier: tests can only
evaluate what the student knows on a given day, not their long term
retention of the material.
Lee, I made an argument that the current system, which is used by private
schools and private institutions across the country, is working correctly.
There is a private testing organization, the College Board, which DOES
administer standardized tests, because SOMEONE seems to think that they
are a good measure of student aptitude. The College Board has an
incentive to make certain that its tests accurately measure the abilities
of the students who take it, and its results are highly regarded in
institutions of higher learning everywhere. However, colleges also
(generally) realize that such testing is not a *sufficient* measure of
student aptitude, and so they also look at the direct evaluations of
teachers in classes.
I don't THINK your proposal is an argument for market failure, and I
presume that you think that there's a serious reparable flaw in the way
the market works, so I presume, by making this argument, you mean to say
that there's a great business opportunity available here for anybody who
wants to start up a credentials organization and do it well. This isn't
obvious to ME, but our market has a great way of betting on such things.
Suffice it to say that *I* wouldn't invest in a business plan like this.
-Dan
-unless you love someone-
-nothing else makes any sense-
e.e. cummings
This archive was generated by hypermail 2.1.5 : Fri Nov 01 2002 - 15:05:26 MST