I thought John Denker gave a fairly complete description of various schemes,
and I think I pretty much agree with his analysis.
One of his schemes was designed to provide maximum information. That is the
scheme I use, although my average scores run more like 65% rather than 50%.
My standard deviation is typically 15-20. This means I have students
getting scores all the way from 25% up to 100%. 60-65% is a C and 65%-70%
is a C+, so my average grade is on the C/C+ borderline... a tad bit of
inflation from a C average, but not much. With reasonable lab work the
average student should easily get a C+ or B- in the course.
I accomplish this test average with a blend of easier questions and a blend
of more difficult questions.
However, I have indeed been thoroughly blasted for this grading scheme both
by students and by faculty outside the sciences. Faculty outside the
sciences say it is demeaning for students to take an exam and get only 65%
correct, even if that is a C+ grade. I retaliate by saying I think it would
be demeaning to get 80% correct and only end up with a C, but many people
just don't see it that way.
In my opinion it still boils down to my professional judgement about what
grades mean. If I write a test and someone gets 65% and I declare... "You
got the basic stuff that a person has to know from taking this course, but
you weren't able to show that you know the material at a level much beyond
that. Therefore I would say your performance is average."... If I am a
Ph.D. physicist with many years of teaching experience and lots of
successful graduates in teaching, research, and other physics-related
jobs... why would anyone find fault with that? Isn't that kind of judgement
what they're paying me the big bucks for?
Michael D. Edmiston, Ph.D. Phone/voice-mail: 419-358-3270
Professor of Chemistry & Physics FAX: 419-358-3323
Chairman, Science Department E-Mail edmiston@bluffton.edu
Bluffton College
280 West College Avenue
Bluffton, OH 45817