Given a recent post on grade inflation, I'm debating the idea of switching
to a norm-referenced grading system. Forget the need to establish criteria
for my grading, as long as I know the distribution of grades that the
university is looking for, I can make it so. I can give a variety of tests
and the top 15% will get and A, the next 25% will get a B, etc. After all,
since a C is suppose to represent the "average" student then that is the
grade most students should earn. So what if the cut off averages are 95 for
an A, 82 for a B, 65 (yes, 65) for a C, and 55 for a D.
Students/parents/legislators/etc. will all understand that the student who
had a 94.9 average really only deserved a B in the course. Or the reverse,
where the cut off for an A starts around an 83 to make sure I have that
"ideal" distribution. (You know we have to watch out for artificial grade
deflation as well.)
So what if the students might not actually learn anything, or that some
students who learned a lot would be given lower grades simply to meet some
pre-established "ideal" grade distribution. As long as the grade
distribution "looks" correct everyone will be happy, right?
Okay, now in all seriousness I do not plan on doing the above and I agree
that grade inflation is a growing issue on many campuses. However, the
notion that we can establish some pre-determined "ideal" distribution seems
unrealistic to me, not to mention potentially harmful to the students. And
here's a situation, this past semester we offered 2 sections of research
methods. Each taught be a different professor. At the start of the
semester, there was a difference of about .4 points in the GPAs of the two
sections (on a 4 point scale). (One professor started recruiting his best
and brightest students from other courses to sign up for his section which
in turn created the GPA difference between the two sections... long story
for another time and I bet you can guess which section I had.) But, no
system of checks and balances set by the administration could account for
something like that happening.
And a quick note on my grading system: I establish before hand that a 90 is
the cut off for an A, 80 for a B, etc. and before I grade any subjectively
scored test/assignment (essays, papers, etc.) I get a picture in my mind of
what I would expect from the "average" student and use that as a baseline
for my grading. (In this regard I am using some norm-referencing, but it is
to an abstract student and not tied specifically to the rather small sample
I have in my course.) I don't even pay attention to the distribution of
grades until I've graded all of them, and many times don't even pay
attention to the distribution until the end of the semester (unless there
seem to be lots of people at one extreme or the other). Sometimes I end up
with grades a lot higher than I expected (my research methods course this
past semester blew me away with their final projects compared to previous
semesters). Other times they turn out lower than expected (the same class
bombed the mid-term.) (I taught the course differently this semester and
the grades on assignments/tests showed it more dramatically than I expected.)
Basically, I would be opposed to any system set by the administration (or
anyone) that tried to establish what the distribution of grades in my
course were suppose to look like.
- Marc
G. Marc Turner, MEd
Lecturer & Head of Computer Operations
Department of Psychology
Southwest Texas State University
San Marcos, TX 78666
phone: (512)245-2526
email: [EMAIL PROTECTED]
