Thoughts on Survey Seminars

I've had a tab open for a while containing an Inside Higher Ed article on a new approach to introducing science at Emory University:

David Lynn, who chairs the department of chemistry at Emory University, spoke about Emory's seminar program for entering freshmen. All Emory freshmen must take a seminar the first semester and the one for math and science teaches students how to think like a scientist.

The course consists of five modules. Each module is taught by a grad student who presents his own research, guiding students through the research process, from designing studies to defending results. For the final, students must write and defend a proposed project.

The article mostly talks about the effect of the course in terms of teaching the grad students how to teach, but I flagged it because I want to say something about how it affects the undergraduates. It's been sitting there un-commented-on for the past week, though, in large part because I've been teaching our own first-year seminar class.

We've been doing something similar for entering students with an interest in physics for eight or ten years, now, and from the undergraduate perspective, it's kind of a mixed bag.

(Continued below the fold.)

The structure of our course is similar in some respects to the Emory class. We invite first-year students who have expressed an interest in physics to take the class, and five faculty each give a series of lectures about a topic relating to their research. I talk about laser cooling, one of the astrophysicists talks about black holes, our chairman talks about nuclear and particle physics and does a Rutherford back-scattering lab using the accelerator.

The transparent goal here is to expose the students to the most interesting and exciting stuff we know about in their first term, to get them interested in physics and astronomy, and hopefully draw them in as majors. We do it this way because the first couple of terms of introductory physics tend to be kind of a hard slog, and we hope that presenting some more interesting material up front will provide an incentive to stick with physics.

As I said, this class has been going on for a while now, and it's kind of a mixed bag. On the positive side, we've seen a dramatic increase in the number of physics majors over that span-- six or eight years ago, it wasn't unusual to have one, two, or even no physics majors in a graduating class. Last year we had eight, this year we have thirteen, next year we expect another ten or twelve. Many of those students entered the department through the seminar class.

On the other hand, though, some of the students have had very negative things to say about the class, mostly due to the pace. We're on a trimester system, with ten-week terms, so five sets of faculty lectures means we each get two weeks. I'm trying to discuss laser cooling, atomic clocks, and Bose-Einstein Condensation in a total of six lectures. We end up throwing a huge amount of material at the students very quickly, and some of them end up feeling overwhelmed. One or two of our majors told me they hated that class, because it moved so quickly.

So, as I said, a mixed bag. I really enjoy giving the laser cooling lectures, but I'm not sure how much the students really retain from them, other than that it's fun to smash stuff with liquid nitrogen. It has been an effective recruiting tool, though-- I get much more interest in summer research jobs from students who have taken that class-- so I wouldn't want to get rid of it, but maybe it needs some tinkering.

The problem is really the ten-week term. The Emory class seems like it probably has more of an overarching theme or structure to it, which would probably help matters, but I'm just not sure we would have time to do that, unless we cut down the number of faculty to a point where we wouldn't be able to present a representative range of topics. It's an interesting idea, but I'm not sure that what w're doing now is broken badly enough to require major repair efforts.