Chronology Current Month Current Thread Current Date
[Year List] [Month List (current year)] [Date Index] [Thread Index] [Thread Prev] [Thread Next] [Date Prev] [Date Next]

Re: Extra Credit (was Where Have All the Boys Gone?)



I have already expressed my viewpoint that allowing/assigning extra credit
to enable a student to raise his/her grade is not a good thing. In this
posting I would like to describe how I grade and why I choose to grade that
way.

(1) I have a grading scale that is pre-established. I do not curve grades.
If everyone truly earns A grades according to my scale, I give all A's. In
22 years I have done that twice (in classes smaller than 20 students). I
also would be willing to fail the whole class. That has never happened, but
several times I did not have any A grades, and once the highest grade was C.
(I'm still talking small classes of 10-15 students.) I did get called to
the Dean's Office when the highest grade was C, but everything was
documented and my grades were upheld. Most often the class average is B- or
C+ and about 10% earn A.

(2) All exams are averaged and constitute 50% of the grade; lab reports are
averaged and constitute 30%; problem sets are averaged and constitute 20%.
I do not have separate mid-term and final exams. I space four exams over
the semester, with the last exam given at the final exam time, but not
counting any more than the other exams. All exams are 2-hour exams.

There is a lot of helping going on when students work problems and labs.
Exams are strictly individual. This means my students have their grades
determined 50% by what they can do by themselves, and 50% by what they can
do with help from other students (or from me).

I do not think everyone has to break their grades down this way, but it
works for me. However, I sure wish everyone would consider what I do under
(1) above. More on that later.

(3) The problem-set problems come from the text, and these are "ranked" by
the author, so I can choose mostly "easy", a few "moderate", and rarely a
"difficult" problem in each problem set. Lab reports are graded carefully
for science and also for grammar and writing style. Therefore, assigning
and grading problem sets and lab reports is not difficult... it just takes
time.

The difficult task is determining the questions for the exams. It would be
easier in a large university with 200 students in the class because "curving
the grades" or adjusting the difficulty of the next exam would be easier.
However, I believe it is not valid to curve the grades in the "normal
manner" if the class size is smaller than about 50 students. This means
small-college teachers like me (class sizes ranging from 10 to 25 students)
and high school physics teachers are probably in the same boat.

The problem arises because the student quality and student attitude
fluctuates so much from year to year. Suppose I give an exam that is
similar to the exam I have given in previous years. Suppose the current
class has a class average of C- whereas earlier classes had an average of B-
on a similar exam. What happened? Is this (a) a bad class, (b) a bad
textbook, (c) I lectured poorly, (d) the exam was too hard or at least had
some unclear things on it? I would also ask similar questions if the
average grade on an exam were A.

Hence, I have to judge the whole class to determine if the next exam needs
to be easier, harder, or the same. Although this is difficult, after
teaching the first 5 to 10 years I got a pretty good feel for what the
"national average" student is like, even though I do not have a "national
average" class in any given year.

Some people say "you are indeed curving the grades." I say I am doing
something different. My first goal is to be consistent from year to year in
the types of questions I ask. However, unless I use old exams over again,
this is difficult to do. Therefore, if I goof up and ask a tricky question,
or make the test too long, or put a really hard question on the test, I need
to have some way to adjust for that. But I can't do it based upon the curve
of any particular class, because any particular class is too small and does
not necessarily have a "national average" composition. We know we don't
have a uniform cross-section of students attending Bluffton College.

High-school teachers might have closer to a uniform cross section of
students, but all my teacher colleagues, and especially my wife, tell me
they see drastic differences in overall class aptitude from year to year. I
think it is at least unfair, and probably not appropriate, to make each
year's physics class come out to a particular average grade if there is any
indication that the quality of the class varies from year to year. The
medium-quality student in a very good class will get a lower grade than s/he
deserves, and the medium-quality student in a poor class will get a higher
grade than s/he deserves. The way to avoid this is to try very hard not to
curve grades to any particular class, but perhaps do some very broad
"curving" over several years to guide you as to whether your tests are too
hard or too easy.

Anyway, the only adjusting I perform is to decide if I need to throw out a
problem, or decrease the weight of a problem, or make the next test slightly
easier or harder... based upon comparison of this class to what I think a
national-average class is like, based upon previous classes I've had, and
also how well those previous students did on normalized tests such as GRE,
MCAT, etc.

(4) When students ask for extra credit, the answer is always no. However, I
will take the time to help the students assess where they are having
difficulty. I will suggest problems they can work on. I will even provide
answers (sometimes worked out by me) on unassigned problems if they elect to
do them. When they ask if they can get extra credit for what they have done
I always say, "You bet... you are going to get extra credit on your next
exam because you have taken the time and expended the effort to understand
the types of things that hurt you on your last exam."


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