While a grad student at MIT, I tought sections of a few courses. This was my first real teaching experience; as an undergrad, I had graded and/or been course assistant for courses, but while that involved helping students in one-on-one situations, it didn't involve teaching classes. This describes my experience with those courses, as best as my (doubtless faulty) memory allows me a year and a half after the fact.
This course is two courses in one: the first half is a very accelerated single-variable calculus course, intended as a sort of review course and the second half starts the multivariable calculus sequence. A professor (Arthur Mattuck) lectured three times a week to all of the students; the students then met in smaller sections twice a week. I taught one of those sections.
The professor was very organized: for each section, he gave us copies of his notes that he'd lectured from and some notes on what he expected us to cover in sections. A typical section involved my doing some initial lecturing, then giving out some problems for the students to work on individually at their desks, stopping them after about 10 minutes and presenting the solutions to the problems on the board, and then lecturing some more. The professor told us what to lecture about and what problems to give the students; typically, the lecture material wasn't new but rather repetition or examples of what had been talked about in class, but there was often some new material in the lectures.
This level of organization was great for my first teaching experience: I had enough to worry about with the mechanics of teaching, without having to worry about course planning on top of that. I'd try to avoid that kind of organization in the future, but it was good for then.
The main thing that I didn't think went particularly well was the seatwork. I gave the students two problems and had them stop working after 10 minutes. They worked individually; they also could ask me for help if they wanted. Most students didn't finish both problems most times; students didn't always even finish one of the problems. I felt rather uncomfortable stopping them while they were in the middle of working, and sometimes gave them more time. One can argue, however, that seatwork can be quite helpful even if they don't solve the problems, because they'll get a lot more out of seeing my solutions after having struggled with the problems themselves, even if those struggles were unsuccessful.
This is a differential equations course. It also meets three times a week in lectures and twice a week in sections; however, it's a much larger course than 18.01a/18.02a.
In the January between the two courses, I'd been to a lecture by Eric Mazur that mentioned some groupwork techniques; I'd also been to a lecture by some MIT professors about that. I'd been curious about this ever since I read No Contest, by Alfie Kohn, so after reading some more on the subject and talking with Lori Breslow, and with the permission of the person organizing the sections for the course, I decided to try to teach my section using groupwork.
I went rather whole-hog with the groupwork: I almost never lectured at all. Instead, after answering questions, I started right in with the groupwork. Thus, I told them to break up into groups, handed each group some problems to work on, and they started working. I wandered around answering questions and monitoring their progress. Towards the beginning of the semester, I tried to end the sections by having students present solutions on the board, but for no particular reason (other than the ever-present lack of time), I stopped doing that after a few weeks.
I initially thought the groupwork was going fairly well. I was suprised at the lack of resistance from the students. The first couple of times, I tried to prompt them to make sure that everybody was following along; after that, I didn't do much overt coaching.
There were some negative sides. In the first place, my students weren't on the average doing any better than those in other sections; if anything, they were doing marginally worse, though I don't think that it was enough of a difference to be significant. So I didn't feel that I was hurting them, and I hoped that, as I got better at using groupwork, my students would start to do better, but it wasn't a ringing endorsement of groupwork. I had somewhat fewer student come to office hours than I expected; maybe they were busy, maybe my expectations were off, but maybe it was a sign of problems in my interactions with them. Also, some students really wanted to see authoritative solutions to the problems.
Another problem was that the groups didn't always function as groups: all too often, one student had a problem with something that somebody else in the group had figured out, and they didn't talk to each other about it, but rather asked me or just didn't say anything. In retrospect, I may have spent too much time wandering around: if I was close to a student, they'd be more likely to ask me their questions than ask another student. But I also should have done a better job monitoring and guiding group interactions. I started to do more of that about two-thirds of the way through the class; it helped.
Over the summer, I taught number theory to a couple of investment bankers. This wasn't an official course; they were just a couple of people who thought that it would be neat to learn some number theory, and tracked me down as a possible teacher.
It was a lot of fun. Not too much in the way of pedagogical experimentation; I tried interspersing a bit of groupwork, but it didn't work, and it was always mostly lecture, anyways. But it was my first time planning a course myself, so that was good practice for me.
Last modified: Wed Jun 28 17:35:39 PDT 2006