Esa M. Rantanen wrote: ...
(1) how do you rate the importance of exploratory and (in particular) graphical analysis of data prior to doing inferential statistics, and
Crucial. Anyone who doesn't do exploratory graphs had better have a visual disability for an excuse.
(2) how do you (or, would) incorporate these aspects of statistics into your teaching?
I use the cruel approach :) Lead them into subtle mistakes in the class, then point them out.
When I teach regression, I get the students excited about the numerical results (F-ratios, p-values, etc). Then I give them Anscombe's data set (http://www.tufts.edu/~gdallal/anscombe.htm ). Once they've chugged through that, and discussed it ("hey, these are almost exactly the same data!"), get them to plot it. Hopefully, an embarassing mistake in the classroom can lead to avoiding this mistake in practice.
Similarly for the CYG OB1 data set, when teaching robust regression. http://www.rss.org.uk/exams/papers/1997/cos_paperIII97.pdf
For teaching boxplots, I give students Darwin's cross vs. self polenated plant data, and get them to try to determine the difference using histograms. Then show them using boxplots.
http://statwww.epfl.ch/davison/SM/SMsample.pdf
The important thing I've found when using this approach is to make sure they're using some sort of software they're comfortable with. If you don't, you'll just get grumpy students who believe they're wasting effort. If the mistakes can be merely embarassing rather than exhausting, it works pretty well.
I share your frustration - in New Zealand, it's possible to get an engineering degree without taking a single stats course. Scary.
Cheers
Jason -- Indigo Industrial Controls Ltd. http://www.indigoindustrial.co.nz 64-21-343-545 [EMAIL PROTECTED]
. . ================================================================= Instructions for joining and leaving this list, remarks about the problem of INAPPROPRIATE MESSAGES, and archives are available at: . http://jse.stat.ncsu.edu/ . =================================================================
