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: [Phys-l] assigning grades



I've used both grading systems that Carl discusses. Let's call the
system in which each category always has the same overall weight (test
average = x%, homework average = y%, etc.) System 1, and the system in
which each graded element has a definite number of points System 2. In
system 2, student grades tend to fluctuate more over the course of the
semester. Student grades creep upward as additional homework grades are
added, and then suddenly drop after an exam, creep up again, and then
drop again. In system 1, the variations in the overall grade are
smaller, because the variations in grades within a category tend to be
less than the differences between grades in different categories.

Daniel Crowe
Oklahoma School of Science and Mathematics
Ardmore Regional Center
dcrowe@sotc.org

-----Original Message-----
From: phys-l-bounces@carnot.physics.buffalo.edu
[mailto:phys-l-bounces@carnot.physics.buffalo.edu] On Behalf Of Carl
Mungan
Sent: Friday, May 12, 2006 2:07 PM

Here's an admininstrative issue I am thinking about: a better system
for determining grades so that a student always knows what their
grade is at that point in the course. (At our school, we are required
to report interim grades just after the one-third and two-thirds
points of the semester.) I welcome comments on what you think about
the pros or cons of the following system.

Assign a base number of points to every graded item. I suggest 1
point per homework assignment, 1.5 points per lab report, 10 points
per in-class test, and 20 points for the final exam. (The number of
points for an item can be adjusted for say a particularly long or
short assignment.) I can now count up how many points a student has
at any time and divide by the maximum possible number of points
available by that date to deduce their grade. This is in contrast to
my present system, where say their test average counts for 45% of
their grade regardless of how many tests there have been up to that
time. (This particularly doesn't make sense early in the semester
when there's only been one test.) In other words, I'm going to weight
each *individual* assignment, test, and lab for a certain number of
points each, rather than weighting their homework, test, and lab
*averages* for a certain number of points each.