One of the most distasteful parts of my job is assigning marks to students. It's a task that's impossible to do right but it's of critical importance for the students. That means it causes problems, no matter what.
It's also a process that students don't generally know a lot about. The point of this page is to try to give students an understanding of what goes on when instructors (in particular, me) assign them a mark. There are no strict rules here, just some thoughts on the process.
The problem is in how your mark is determined. Here are the three marks that I think it makes sense to talk about:
- The mark you deserve.
The ideal. The platonic form. The mark that would be handed down by an omnipotent being. This mark would be determined by evaluating everything you know about the subject at hand.
The problem is, it's hard to figure this out. Giving assignments and tests help approximate this as closely as possible. This is why I consider creating tests to be the hardest part of my job. I'm trying to evaluate everything and make the outcome match the mark students deserve. And, it has to be done under a time limit.
This is, by the way, the reason my exams tend to be a little on the long side for some courses. It's the urge to pack in as much as possible to have a broader evaluation to work from.
Bad tests and assignments are the ones that don't approximate the ideal mark well. If I gave a test with a bunch of questions like "One or two? Guess.", it would come out with a nice neat bell curve and predictable average. The problem is, it doesn't tell me anything about the mark you deserve.
- The mark you earned.
Estimating your ideal grade is too hard and too subjective. The only way to be anywhere close to fair is to come up with some objective way to determine a grade.
So, we give out tests and assignments and you do them to demonstrate how much you know. After you've done all of the tests and assignments in a course, you've earned some marks. They are totaled up and converted to your grade. Realistically, this is what we have to aim for at the end of the semester.
If the tests and assignments have been designed well, this will be a good approximation of the mark you deserve. Either way, this is the mark you should get; setting the tests and assignments is the instructor's job and he or she does it, for better or worse.
This is part of the reason that saying "I knew the answer for this question, but just didn't write it" won't get you anywhere. We're evaluating what you did, not what you know. It might not be the ideal thing to do, but it's the best thing that actually works.
- The mark you got.
When you get your mark from the Registrar's Office, this is what you get.
The process of marking and recording marks is not perfect. It's done by people who, despite being generally skilled, make mistakes. These mistakes are the difference between what you earned and what you got. This is why you are allowed to review your final exam and should look back at the other marked work in a course.
This is why the marking questions are usually painful for all concerned. Instructors are concerned with giving out the mark you earned by making sure the marking was fair (and hoping that it's the mark you deserve); students are generally hoping for higher marks. When those ideas come into conflict, it causes tension.
Basically, it's the instructor's job to make the ideal marks line up as closely as possible with the marks students earn. It's part of the job and isn't generally up for discussion. If nothing else, it's impossible for a student to really evaluate the mark they "deserve". It's a frame of reference problem.
Making the mark you got match up with the mark you earned is part of the marking process. This should be reviewed by students and fixed when there are problems.
The real job of a student is to make sure the number of marks they earn is as high as possible. Figuring out how to do that is what being a student is all about.
Other Factors
In the quest to get the mark you earned to match the mark you deserve, the design of tests and assignments is the biggest factor--if an exam is truly badly written, there isn't much you can do about it.
After the work is assigned and done, there are still several variables that affect the final mark that's given.
- Marking.
Marking is hard.
There are a lot of choices to be made along the way. How many part marks should be given for this? Does this student really know what they're saying, or just trying to fake their way through it? Often, answers can differ by just a few words, but those words indicate vastly more or less understanding of the topic.
Everyone makes these choices differently, it's just the nature of the problem. Usually, as long as the choices are made reasonably and (above all) consistently, everything works out well.
Marking a particular piece of work easier or harder will affect the marks given. It's all part of trying to make the mark you earned match the ideal.
- Curving and scaling.
Curving or scaling marks is a greatly misunderstood thing.
As far as I'm concerned, it's just another way to make the "earned" mark line up with the ideal mark that you deserve. If the marked work in the course has been too easy or hard overall, then the final calculated marks won't line up with the marks students deserve.
Since we only give letter grades at SFU, the easiest way to move things around is to adjust the numeric ranges that correspond to each letter grade. How much or little the ranges are adjusted is determined by the instructor's ideas about what marks students deserve.
There is also the problem of getting marks accepted--after an instructor submits final marks for a course, they must be approved by the department chair. The idea is that sets of marks that are too high or low should be rejected and reevaluated. My experience is that this is a fairly minor factor. If reasonable attempts are made to closely approximate the marks students deserve, they will be accepted.
John Hebron has some notes on curving marks that match my thinking about the subject. Also, note the page on the bell curve. Note that in the curve for the "Exam That's Too Easy", there isn't much to distinguish students because they're all clustered above 80%. This is why instructors generally err on the side of making an exam too hard. If it's too easy, it doesn't tell you anything; if it's too hard, you can always scale the final marks up.