home *** CD-ROM | disk | FTP | other *** search
- Comments: Gated by NETNEWS@AUVM.AMERICAN.EDU
- Path: sparky!uunet!spool.mu.edu!darwin.sura.net!paladin.american.edu!auvm!WUBIOS.WUSTL.EDU!PHIL
- Return-Path: <phil@wubios>
- X-Mailer: ELM [version 2.4 PL17]
- Mime-Version: 1.0
- Content-Type: text/plain; charset=US-ASCII
- Content-Transfer-Encoding: 7bit
- Content-Length: 3098
- Message-ID: <9212221523.AA06586@wubios>
- Newsgroups: bit.listserv.stat-l
- Date: Tue, 22 Dec 1992 09:23:17 -0600
- Sender: STATISTICAL CONSULTING <STAT-L@MCGILL1.BITNET>
- From: "J. Philip Miller" <phil@WUBIOS.WUSTL.EDU>
- Subject: Re: Tutorial Software (fwd)
- Lines: 61
-
- Forwarded message:
- dkw4@NIOSHE2.EM.CDC.GOV Deanna Wild writes:
- >
- > I forwarded J. Philip Miller's comments about the dangers of using
- > spreadsheet packages with ill-conditioned data to some of the industrial
- > hygienists I work with. What follows is a response from one of them.
- >
- > Dr. Miller, could you answer these questions directly if you've got the
- > time/energy? If not, I'll take my own best stab 'cause I tend to concur
- > with your opinion. I'd also be happy to hear from anyone else on this.
- >
-
- I suggest anyone really interested in this should go back to the archives of
- STAT-L & EDSTAT-L and get all of the debates. I can't say the basic concerns
- any better than Paul Vellman who wrote to EDSTAT-L a little over a year ago:
-
- o / o / o / o /
- --Cut-here-------X---------------X---------------X---------------X----
- o \ o \ o \ o \
-
- I'm sorry, Bob, but you persist in ignoring what several of us have been saying
- on this board and on Stat-L. The formulas in the textbook *must not* be used
- for computing. It doesn't really matter whether they are used in a spreadsheet
- or somewhere else. It is wrong, dishonest, and frankly, incompetent to teach
- students to use these formulas for computing. You are teaching them false
- information (or worse, information that seems to be true and is almost true
- most of the time, but fails without warning.) Would you teach that the Earth
- was the center of the universe simply because you have not seen with your own
- eyes clear evidence to contradict that "obvious" fact? Or would you trust that
- experts in the field who have studied the matter might have some knowledge that
- you should convey to your students? A Geocentric universe works pretty well
- until you push the model; the Sun, Moon, and stars circle the earth and all
- looks OK. Well, we do have occasional wanderers in the sky, but we can
- "reformulate" the problem a bit and account for them most of the time...
-
- Several experts in statistical computing have said during the Stat-L discussion
- that the choice of algorithm is an important one and that it is wrong to teach
- students otherwise. There are many well-written statistics packages that do the
- computing correctly, and some of them are good teaching tools. Given all of
- this, there is simply no excuse for teaching students to calculate statistics
- with unstable algorithms on inappropriate software platforms.
-
- (For those of you who missed the several months of earlier discussion on
- Stat-L, I appologize if I seem a bit peeved. We have been over this ground
- several times but seem to make no headway.)
-
- o / o / o / o /
- --Cut-here-------X---------------X---------------X---------------X----
- o \ o \ o \ o \
-
-
- > Thanks very much -
- >
- > Deanna Wild
- >
-
-
- --
- J. Philip Miller, Professor, Division of Biostatistics, Box 8067
- Washington University Medical School, St. Louis MO 63110
- phil@wubios.WUstl.edu - Internet (314) 362-3617 [362-2694(FAX)]
-