Hi Andy and Geoff:

My experience is that the idea of kriging is pretty
easy to get across, in fact most classes can come up
with it on their own ("How about a weighted average
with higher weights on nearby observations?")  The
details on how to choose the "best" weights tends
to bog people down.

I talked to Steve Rathbun at Georgia yesterday and he's
teaching an environmental statistics class to ecologists,
and just covered the variogram.  They just had a lab on
variogram fitting yesterday.  He said some got the idea
right away, others were pretty confused.  "Fitting by eye"
works pretty well for giving them the idea, so you can
do a lab on kriging where it works.

The issue you're likely to run into is combining a new,
strange thing (variogram fitting) *and* learning a separate
software package to do it.  (You get the "I know what I want
to do, but don't know how to get this (#*$&@(&ing package
to do it!" syndrome).  If GeoEAS is simple enough, you may
not have this problem, but "simple enough" is relative to
the student's background.

At least that's been my experience.

Lance