NetNews Usenet Archive 1992 #27/NN_1992_27.iso/spool/sci/psycholo/4470

Text File | 1992-11-21 | 2.7 KB | 55 lines

Newsgroups: sci.psychology
Path: sparky!uunet!cs.utexas.edu!zaphod.mps.ohio-state.edu!moe.ksu.ksu.edu!ux1.cso.uiuc.edu!news.cso.uiuc.edu!s.psych.uiuc.edu!ffujita
From: [e-mail] (Frank Fujita)
Subject: Re: Sanity Certification
References: <[e-mail]>
Message-ID: <[e-mail]>
Sender: [e-mail] (Net Noise owner)
Organization: University of Illinois at Urbana
Date: Sat, 21 Nov 1992 18:46:58 GMT
Lines: 43

[e-mail] (Mark Robert Thorson) writes:
>It seems to me that it is possible to create a first-level screening
>test for insanity, e.g. combining elements from the MMPI, Rorschach,
>Stanford-Binet, electroencephalogram, etc. [brave new world proposals
>deleted]

Well, someone else will question your motives.  I'd just like to point
out some of the technical difficulties, and let others discuss ethics.
First, there is the problem of low baserates.  Second is the problems of
cutting scores.  Third is the problem of the changing definition of
insanity coupled with the lag time in validation of tests.

First: Low Baserates.  Because very few people are insane (less than 5%)
any test that will try to predict who is insane will have to be *very*
accurate.  (This problem in *not* domain based -- the AIDS test has the
same problem.  In low baserate populations, there are more false-
positives than correct hits).

Second:  Cutting scores.  No matter how *good* your insanity test, there
will be a distribution of scores, and you will have to decide (someone
will have to decide) that a score of 19 is sane, and a score of 20 is
insane.  (Assuming that a 3 is clearly sane and a score of 40 is clearly
insane, there is still going to have to be a cutting score).  Thus,
while you can keep out the worst people (as defined on the test) you
will either have to exclude *many* sane people from those occupations
(with a low cutting score) or include many insane people.  In either
case, your cutting score will be arbitrary, and some people will be
discriminated against -- Like if they kept people with a 19 ACT out of
college but let people witha 20 ACT in.

Third problem: Changing definitions of insanity.  It takes about 10
years to develop and validate a test to the level where it can be used
in individual prediction.  However thei definitions of insanity change
about every 5 years.  At any given time, we would only be able to
identify people that would have been considered insane a while back, and
anybody so identified could reasonably argue that under current
guidelines they weren't insane.

I took your post seriously, but others will probably (rightly?) dismiss
it out of hand as fascist, or communist, or something depending on their
political orientation.

Frank Fujita