A Professor’s Thoughts on Grading

August 3rd, 2009

I’ve never been a big fan of being graded or grading others.   Ideally we should all learn for the sake of learning, but I have to admit that grading, from a number to a letter to a gold star can be a stupid but effect motivator.   Moreover, the grades are used by admissions committees for college, grad school, and taken into consideration by employers.

First of all, I think some subjects shouldn’t be graded except as pass/fail, as these subjects are very subjective.   Art, for instance.   I took drawing in college, and took it pass/fail, as I thought that was fair.   I would have gotten an A (I was pretty good in the day) but it wasn’t important.   Same thing with creative writing — I would have gotten an A, but took it pass/fail.   All I needed was a dick prof who didn’t care for science fiction who decided it was his duty to help his students see the light, so to speak.

Any course where the so-called “BS” works should also be pass/fail.   I had a college roommate, a smart guy, who was nearly a straight-A student and took lots of humanities courses.   He had a system.   He wrote his own thoughts and opinions on his first paper, and if he got anything less than an A he would change tactics and regurgitate the attitudes and opinions of the prof.   Worked every time.   I just shuddered when he told me stories about this.

I believe that those big intro courses, often featuring basic information and multiple-choice tests, should not be graded based on a curve.   They should be easy enough that most students working hard can get an A or B, and only the real slackers scoring less.   Often these courses, however, by virtue of their sheer size, are the ones that most likely end up with normal distributions of grades.   When I taught intro astronomy, mine was a curve skewed to higher grades.

Upper-division major courses, and grad courses, ought to just have grades of A, B, and C, ideally.   Everyone at that level should love their subject and not be too bad at it, although disasters sometimes happen at every level.   Grades of C are pretty much regarded as failing at the grad level, and sometimes those courses must be retaken.

I think grading is easier and fairer for quantitative disciplines, like math, science, and engineering.   Everyone can see the numbers, and the scales can be defined at the start of the course.   There should be little to no confusion.   When I started teaching grad courses, I initially gave lots of research-type problems that did not lend themselves well to simple numeric scoring.   I have kept these, but added problem sets and exams that are of the same type we use for our qualifying exams which every grad student must pass to move on to PhD Candidacy.

[Added later in response to a good point.   Professors do screw up either covering material, conveying expectations of material on exams, or just plain write stupid hard exams from time to time.   Curved grading is an appropriate response to these events, and I put this exception down in my syllabus, explicitly stating that it is there to work in a student’s favor and protect them.   Ideally an experienced professor writes fair tests, but the world is not ideal, unfortunately.]

Students should be aware that professor’s grade distributions are evaluated regularly and suggestions are made if they’re higher than average, so even profs get graded.   Our dean likes to see low grades and high student satisfaction, as he regards that as being the signature of good teaching.   Sometimes easy grading leads to inflated teaching evaluations by the students, although I’ve seen research indicating the strongest correlation is with the student’s expectation of their grade going into the final (evaluations are done before final grades are given).

I’ve had a year off from teaching, and will resume my duties later this month with graduate cosmology.   That course I’ve taught before, but will have a new one to teach in the spring for freshman/sophomore astronomy majors and am giving that one some thought already.   There’s a lab component, so that gets factored in, too.

Anyway, it’s a necessary evil, grading, and super important to the bean counters who must know how things are going.   It’s a strange thing, though, that in real life and even real academic life (e.g., publishing in professional journals) we stop getting grades.   Somehow everything works anyway, and people strive to do a good job.   Why can’t that be the case in school?   At least with major courses?

Share/Bookmark

You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.