Yesterday I graded thirty midterm exams from my Art of Film course. I actually enjoyed composing this test, by not only incorporating material from the textbook (and the book's online study guide) and the lectures, but also frame captures (off the DVD) from the films that we've screened this term. Test design is an artform that I've only learned from the exams I've taken during my long life as a student; I don't think I've ever had a lesson in how to make a test, though in graduate school there were many collaborative moments where I would brainstorm test questions with my advisors or colleagues. Being trained predominantly in literature, too, means that most of my background in exams is in essay exams.

I'm walking down memory lane because I found myself revising my test as I graded it. Students are experts at finding loopholes: for example, earlier in the term, a student took issue with my use of the phrase "face-to-face" in a true/false question. As a form of film identification, I'd asked whether or not a character comes face-to-face with a bear in both Chaplin's The Gold Rush and Keaton's The General. The answer was true -- both films feature surprise encounters with a grizzly. But the student argued that the characters don't really engage the bear "face to face" (i.e., they don't bump noses). I saw the logic of her point and the flaw in my use of an idiomatic expression. So I accepted every answer in the test that had, essentially, said a true statement was false. And yesterday I found myself accepting creative answers to exam questions, like fill in the blanks, which produced a wide range of answers that I didn't expect, some of which were technically correct, so I accepted them.

Accepting multiple answers and changing the test based on them must be a common experience, of course. Part of the problem is my lack of experience in this area; the other is the way that students are smart at test-taking and applying creativity when the stakes are high, when they probably should have put that energy into studying instead. But the point I want to raise is not so much about test design or student performance and evaluation as it is about teaching as revision. This concept is one I've been thinking about a lot lately -- how the act of teaching teaches the teacher. Changing my test as I graded was a form of rethinking the approach to class material, researching how I will teach it, and contemplating how I will assess it in the future. I now see grading as a form of revision of the course material, not simply a matter of student assessment or coaching. And while this is always the case to some degree (for example, if students in my composition course are all using fragments, I might change the calendar to teach the craft of writing in complete sentences), I nevertheless haven't really thought of the process so consciously as I have this weekend, grading the first "big" test of a newly prepared course. It is important for teachers to remember that class material is always a work in progress, even when the class curriculum becomes routine and repetitive as the years go by.

You can ping this entry by using .