Warning! Steep grade ahead!

So . . . it's grading time!  That bittersweet time of the year where I'm happy that the semester is just about over, but faced with a mountain of student papers to assess.  This is a bigger task than most people realize, because I actually have to read the darn things.  And reading them can be tough.  Not to sound trite, but the quality of student writing is usually horrible.  But with one final push, I'm done the semester, on holidays, and ready to face a new batch of students in January.  Even so, there is one change I would like to see with regard to grading, discussed after the jump.

Now, don't get me wrong.  I have enough experience in assigning grades to be able to know the difference between a 77 paper and a 78 (I have been doing that for about five years, so I'm no expert, but I do have solid handle on it).  I am not saying that grading on a 100-point scale is arbitrary or meaningless.  I do it every year, and I stand by the grades I have assigned.  I just think there might be a better way.

At the beginning of the semester I wrote about the need for university classes to better reflect reality.  Students will need to know how to work a certain way when they graduate, and having classes and exams revolve around theory and facts (that are easily Googled or looked up in the textbook) rather than concepts and skills (which will actually be needed, and should be practiced prior to a career), does not make sense.  Grades should reflect reality too.

When you're in a job, what sort of feedback do you receive about your work (if any)?  It's not a grade of  ou performance from one to one hundred.  Even the most detailed feedback, during an annual performance review, is for most companies on a 5 (or so) point scale. But I'm not even thinking about that - I'm thinking about ongoing feedback.  And from my own experience, and that of those I know, it tends to be either a) the work is not good enough (feedback usually provided); b) the work is good enough (feedback usually not provided); and c) the work is excellent (depending on how excellent, feedback might be provided).  Translated into grades, we could call that fail, pass, distinction.  And, to me, that's what makes sense for assessing students.

So while I can tell the difference between a 77 and 78, that difference would be meaningless in a workplace context.  What matters is if the students do well enough overall.  We already use a pass, fail, distinction model for overall student assessment; students either graduate, don't, or graduate with honours (which at many university entails a pretty low hurdle to surpass).  The entire grading model is ultimately pass/fail, because if you get more than 50% in a course you pass, and if you get less you fail.

(Why 50%?  Seems arbitrary.  Do we want doctors who only knew, in university, half of what they were supposed to?  Or even less, for I've heard that it's very hard to fail out of medical school, even if you get less than 50%?  This either says something about the compentency of doctors or something about medical schools, whether it is that the curriculum is irrelevant or the assessment too harsh.)

I know that my suggested grading model will not be popular.  Some professors may like it (though it seems like less work, it really isn't; the work is not in the assigning of the grade but rather the assessing of the paper and writing comments) but administrations won't, because it's harder to rank students and have grade variability.  Students, I think, would hate it.  Having a number grade that can be compared to that of one's classmates is of value to most students.  I have had countless students come to me concerned about their particular grade (sometimes on an assignment worth 5%, where small grade difference will not impact their overall grade), so it's clearly important to them. 

I'll continue assigning number grades for as long as I need to, but I'm not sure whose advantage this is to.  I know that for some types of exams and tests, number grades make sense (e.g. multiple choice, short answer).  But for more involved questions, where decisions or arguments need to be made, it is not the best tool for judging student performance.  And if you're one of my students reading this, don't think you can use this to argue for a better grade - I know that your work is worth that 64.5%.