Since finals are nearly upon us here (and since I'm not quite ready to face the next stack of papers that needs grading), I got to wondering how other academics feel about when the final exam ought to be written and why.
So, a quick poll:
A relevant question, I suppose, is whether what you think is the best time to write the exam from a pedagogical standpoint corresponds to when you actually get it written (owing to your other responsibilities, the unidirectional flow of time, and so forth). Lurking in here are also some worries about "teaching to the test" rather than having an exam that functions as a tool to determine how well your students learned the stuff that you believe it would be good and useful to know even if they weren't being tested on it.
What's your philosophy on final exams, and how does the fact that you have to give final exams influence your teaching leading up to them? (If you're someone who teaches but doesn't have to give final exams, tell us about how that works out.)
I am a lazy person. This means I think and plan to do things in the most efficient and effective way so I have more time to goof off. I write the final throughout the course. so when time to print it out comes, I add maybe one question and give it to the secretary.
This is probably just a rationalization of my procrastination, but I don't think I would want to be thinking throughout the semester of whether I'm covering everything that will be on the final. So perhaps it's actually good that I leave it until the end, and then try to devise a final that tests whatever I've covered.
My final exam and my review session (when I have one) are unrelated. I tend to make the review session student-driven: they ask the questions, and I'll answer them as best as I can. If they happen to ask questions which duplicate exam questions, more power to them---I'll give them what I take to be an 'A' answer (and I will be unsurprised when I don't see that many 'A' responses on the actual exam). Sometimes I have a final already written (or recycled from previous years); more likely, I'm writing the exam at the last minute.
I write it first (8th grade science) but I'm a big believer in Understanding by Design. It helps me focus on what's important when I'm teaching rather than bogging down in minutiae.
I write my exams after the review sessions, because students contribute ideas for the final essay questions. I hope that helps them put together some ideas about what we've been working through all semester. And then it helps them study in a focused way. (There are more essay questions to study for than will be on the final, but helping to put them together and thinking about them gets students started on studying. I hope so, anyway.)
My review sessions are brainstorming sessions, and focused on the students putting together what they've learned.
and blood cell cookies:
More internet silliness
Oddly for someone who is the living embodiment of the second law of thermodynamics, I write my exams in the second week of first term. Not through any particular organisation on my part but because we have to send the papers out for approval by external examiners.
Fortunately, although teaching to exam is possible, I generally don't remember the questions and have to remind myself half-way through the term. I cannot vouch for whether or not my colleagues are equally principled. Having said that, setting the questions before the lectures are prepared does give a little impetus to change some of the material, which is useful.