On Writing Exams

I’m writing this little post as I take a brief break from the pleasure that is known as “writing an exam.”

I really love teaching. I always wanted to be a teacher. I like getting in front of a classroom and trying to impart upon my students not just the knowledge that’s rattling around in my head, but also my enthusiasm for the subject. I really love what I do, what I study. I want everybody to know what I know.

Unfortunately, there are these hurdles. We can’t just dance about merrily in front of the classroom day after day (though I do dance as often as I can.) We are obligated to come up with ways to assess our students’ gain of knowledge. We can do exercises, take-home quizzes, laboratories and so on, but the traditional means of assessment is to give regular exams. When I was a student – before I ever sat on the other side of the front desk – I had no idea how difficult of a task that was. It is among the most difficult things I have to do as an instructor, to write an exam that will fairly assess what my students have learned.

Writing good exam questions is hard! First, you have to decide on a format. Will it be multiple choice? Short answer/essay? True/false? Should I have the students write a novel in a Blue Book in one hour? Then, you have to develop the questions. You’d think that multiple choice would be easy to write. It’s completely mindless to grade, of course. Multiple choice questions are actually the hardest to write, because it’s really easy to write a question where more than one answer would be correct. And there’s a terrible down-side for students: There is no possibility for partial credit.

My exams are almost entirely short answer/essay, with a smattering of vocabulary tossed in. These questions are a little easier to write, but much harder to grade. And if you’re not careful with your phrasing, you might inadvertently ask a question completely different from the one you thought you were asking. But the bright side is that students can get partial credit. If I can see where a student got off track, I can give credit for the bits he or she got right. And, if a question was really badly written, it becomes obvious right away and I can account for that before returning the exams.

Right now, I’m doing the one thing I usually don’t do until after the exam is administered. I’m writing the key. I do always prepare a key, and if I can do it in advance it’s a good thing. I can often catch mistakes before the students trip over them. It’s hard to correct these things after the fact.

A key also provides me a framework from which I can grade. I can determine what parts of answers are the most important, and what the points breakdown will be. I can define key words (which can really speed up the grading process). Also, by having a detailed key, I can get other people to help me grade. This year I actually have a TA, which is great! In years past, I’ve been on my own. Trust me. Grading exams is no fun!

The annoyance of writing a key is that I pretty-much have to take the exam. I’ve just spent hours writing it, now I have to take an hour answering it. Later, I’ll spend multiple hours grading it. Trust me, students. You’ve got the easy end of this deal.

And speaking of which. I should probably finish writing this key. Or… maybe I’ll finish it on Friday when I proctor the exam. Hmm. I like that plan.

**** UPDATE (the next day)****

Well, five of 24 students have had the exam now. I’ve decided I don’t like how the points break-down is on the exam. In fact, I really hate this exam. Something just bothers me about it. It will come clear when I grade them, I suppose. It’s too late to change anything now.