----- Forwarded message from Frank E Harrell Jr -----
I'd like to make a somewhat related point. There are many educational
tools that I've found have a great effect on non-statisticians. One if these
is to take one of their datasets, randomly permute the column of Y-values,
go through their data mining procedure, and see what it finds. The more
that it finds, the more the client becomes properly afraid of the technique
and respectful of the statistician's careful approach. -Frank Harrell
----- End of forwarded message from Frank E Harrell Jr -----
That's a nice example, though I would not have the confidence that
they would not see it as a wonderful way to discover even more
"relationships"!-) You could also try sorting the X and Y columns
independently to boost R^2. Some of my students just supplied another
example. I was ill and emailed class cancellation well in advance.
Some of these folks seem to practice procrastination as a religion,
and put off the experimental design assignment due the day I was out
and did it with the time series assignment due at the next class. As
a result, some of them introduced a "trend" variable into the
experimental design data, and got a "significant" p-value. I have not
had the opportunity yet to ask them what "trend" measures in this
context, or what value I should plug in for it if I want to make a
prediction of sales when commission is 5% in Division C.
I used to have a sheet with four residual plots per side that I asked
students to interpret on exams. One plot was a big smiley face. The
answer I hoped for was, "There seems to be a pattern here, but I don't
think any of the techniques we have studied would be appropriate to
deal with it." I wonder what the data mining software would do with
it? Smile back maybe?
_
| | Robert W. Hayden
| | Department of Mathematics
/ | Plymouth State College MSC#29
| | Plymouth, New Hampshire 03264 USA
| * | Rural Route 1, Box 10
/ | Ashland, NH 03217-9702
| ) (603) 968-9914 (home)
L_____/ [EMAIL PROTECTED]
fax (603) 535-2943 (work)
===========================================================================
This list is open to everyone. Occasionally, less thoughtful
people send inappropriate messages. Please DO NOT COMPLAIN TO
THE POSTMASTER about these messages because the postmaster has no
way of controlling them, and excessive complaints will result in
termination of the list.
For information about this list, including information about the
problem of inappropriate messages and information about how to
unsubscribe, please see the web page at
http://jse.stat.ncsu.edu/
===========================================================================
