home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: bit.listserv.psycgrad
- Path: sparky!uunet!zaphod.mps.ohio-state.edu!moe.ksu.ksu.edu!ux1.cso.uiuc.edu!news.cso.uiuc.edu!s.psych.uiuc.edu!amead
- From: amead@s.psych.uiuc.edu (Alan Mead)
- Subject: Re: (no subject given)
- References: <PSYCGRAD%92081310101180@UOTTAWA.BITNET>
- Message-ID: <Bsynru.77y@news.cso.uiuc.edu>
- Sender: usenet@news.cso.uiuc.edu (Net Noise owner)
- Organization: UIUC Department of Psychology
- Date: Fri, 14 Aug 1992 06:38:18 GMT
- Lines: 26
-
- Gerri Oppedisano <gerri@WATSON.IBM.COM> writes:
-
- >Why don't you (the advocates of the GRE) run studies on how well GRE scores
- >correlate with graduate grades?
-
- #include <sense of humor>
-
- How would you design such a study? Here's my patented formula:
-
- Select 100 schools at random. Admit everyone who applies. Wait until
- these students retire or die. Compute a criterion (or criteria if you
- wrote that into the contract). Compute correlation.
-
- Send grant money to:
-
- Alan D Mead
- 603 E Daniels
- Champaign, IL 61802
-
- (And if your study design isn't as good as mine, I'm not going to okay
- your grant when its sent to me, as an expert in testing, to review.)
-
- As an aside, Fischer or someone actually convinced some organization
- (the Air Force, I believe) to do the above study for a training
- school. Wow.
-
-