I’ve just (finally) finished writing the third exam for the cyberporn class. I hate writing and giving multiple-choice exams. (Ironically, I don’t mind taking them, but I don’t really get much opportunity to do that.) Here are some of the reasons:
They test superficial knowledge.
You can get them to probe deeper knowledge and comparisons, but such items are really, really difficult to write. It’s easy to find out if students know who started Esquire, but it is a lot harder to write a question that can tell me if they have an understanding of the political and cultural context in which this occurred. In a blue book exam, I can write “What was the political and economic context in which… [etc.].” And then, students actually get to demonstrate not that they can identify a more correct response when presented with it, but that they can think and write intelligently about the matter.
They encourage superficial knowledge.
Because they test superficial knowledge, this is what students try to get out of the lectures. Can’t blame them. Messed up, though.
They encourage students to think quantitatively about learning.
In an essay exam, the different between an A and a B is subjective, and really that makes sense. Is it true that someone that gets 90% rather than 80% on a multiple-choice exam really knows a specific amount more. I don’t just mean that the questions sample the possible space of learned material — I have little difficulty with sampling. What I mean is that the student who forgets a piece of vocabulary, but who could actually say something intelligent about the material, receives a lower score. And he or she then gets a lower grade and thinks that they are somehow “dumber.” The really bright ones catch on and realize that the grades don’t reflect their knowledge, and many of these manage to fake it and train themselves to learn to the test. But many actually believe that they are not very bright because that is what the test says.
I should note that few actually think the final grade in the class represents what the student knows about the material. As faculty we complain that students see the grade as something to be “won” or “granted.” Unfortunately multiple-choice exams feed into this mentality.
They are hard to write.
I know: whine, whine, whine. Bubble tests don’t have to be graded, so what’s the fuss. It takes at least 10 times the amount of effort to write a decent multiple-choice exam as it does to assign a paper or create a blue-book exam. Yes, I know it takes a lot longer (approaching infinitely longer) to grade, but at least that grading is telling you something about the students and what they know.
Cheating is easier.
I do a lot to try to stop this, but I already suspect a pair of people in this class of cheating. Hard to catch them red-handed, though, since the amount of information that needs to be conveyed is so small and discrete. A lot harder to cheat on an essay exam, particularly when it is open book.
But, despite all of this, until they assign me more than a single TA for a class of 400, and until they consider teaching as a serious part of the promotion process, multiple-choice exams are going to be where it’s at.