Teaching Ambiguity and the Scientific Method

As a sort of follow-on from yesterday's post, thinking about the issues involved reminded me of a couple of browser tabs that I've had open for a while, namely this story about an education session at the AAAS meeting, and this Inside Higher Ed article on "Teaching Ambiguity". From the IHE piece:

Tidy may be comforting, but it is also banal, boring, conventional and unrealistic. That's why I have been asking faculty to infuse their classes with the element of surprise. By surprise I do not mean mysteriously taking a rabbit out of a hat, but rather incorporating experimental, untidy open-ended exercises in their classes.

This request is not an arbitrary one. To the contrary, it germinates from a belief that the liberal arts and sciences, and the students who take such courses, often thrive by appreciating complex questions that do not have easy answers. Precisely because students can retrieve facts instantaneously at their finger tips, I am asking faculty to revise their syllabuses to discuss and, yes, teach, ambiguity.

The article, like most pieces in academic magazines, is fairly humanities-centric, , and as usual for these things my initial reaction was that it sounds great, but would be miserable to implement in our classes because of the more hierarchical and service-oriented nature of our intro classes. That is, while the idea sounds fascinating, when I think about trying to put this sort of thing into practice, I quickly run up against two major problems: first, that we have certain material that we need to cover for our own courses, and second that we have certain material that other people expect us to cover when their students take our classes.

When we teach introductory mechanics, as I am this term, we have a fairly rigid syllabus that we need to follow, because we need students leaving the class to know certain things, in order for them to be able to take introductory E&M the next term. And we have a rigid syllabus for E&M as well, because we need them to know certain things before taking intro modern physics, and so on.

And even if we could get everybody in the department to agree to shuffle our course sequence around (which is highly doubtful), we also have the problem that the vast majority of the students taking introductory mechanics and E&M are not future physics majors, but are prospective engineers. Which means that there is an additional expectation that certain content will be covered, coming from the engineering departments. And they're constrained by their national certification agencies, and so on.

I really like the idea of doing more open-ended questions, but that necessarily means being somewhat flexible about the content to be covered-- if students are going to formulate their own questions and investigate them in a meaningful fashion, you need to accept that you might not get to everything in the usual syllabus. And every time I think about it, I end up deciding that we just don't have the flexibility. Ambiguity is great, and relatively easy to apply to disciplines that are a little more fluid, but science is much more hierarchical, and earlier courses are constrained by the need to prepare students for later courses in much more specific ways than you get in the humanities and social sciences.

Which is one of the reasons that I was very interested to see what the AAAS panelists had to say, because the session's focus on teaching science by focusing on scientific methods sounds like exactly the sort of open-ended kind of thing that I always end up deciding that I can't do.

The session started off with a talk by Jon Miller on his surveys of public knowledge, but the real pivot of the session was Judy Scotchmoor's talk about the Understanding Science project, funded by the NSF to develop resources and methods for discussing how science is done, and integrating that with science education at different levels. One of their main tools is a flow chart that replaces the hoary old hypothesis-confirmation-replication description of a single Scientific Method with a much more complicated and fluid diagram expressing more of how the process of real research works. They talk a lot about using this to illustrate the real progress of science, and how it goes back and forth between different levels and ideas. They have a very detailed illustration of this using the example of Walter Alvarez coming up with the idea that an asteroid killed the dinosaurs.

I had heard Scotchmoor give more or less the same talk last year, I think at the March Meeting, so that wasn't too new, but there were two other talks promising to talk about how this is done in practice. One was Karen Oates for WPI, who talked about organizing bio/med classes around public health sort of issues-- rather than offering an abstract course on cell biology, for example, doing a course on cancer, which ends up bringing in most of the major issues in cell biology. The other was Mark Stefanski, who teaches high school biology in California, and has his students devise and carry out their own experiments to get a better understanding of how the scientific process works (using Scotchmoor's flow chart among other resources).

So, what did this tell me about implementing ambiguity in the science classroom? Sadly, not much useful. Don't get me wrong-- the classes they described sounded great. But they had basically nothing to offer on the issue of finding a way to balance open-ended inquiry with specific content. It was sort of buried in the talk, but Oates mentioned briefly that her cell-biology-by-talking-about-cancer course was not a major-sequence class. And Stefanski was asked directly about how he managed to fit his course into the current mania for high-stakes testing and state standards, and said, basically, "I'm lucky enough to be at a private school where we don't have to worry about that sort of thing." Which is great for him and his students, but not all that helpful to anybody else.

So, I'm stuck with the same situation as usual: I like the idea, but it seems to work best for courses that aren't the courses we mostly teach. We try to cram a few inquiry-based labs into the intro courses, but it's tough to manage with the time and content constraints that we operate under (which are greater than most schools, because we're trying to fit almost a full semester's worth of material into a ten-week trimester). And while I'll certainly look at ways to use these ideas in the non-major seminar course I'll be teaching next winter, I have a hard time seeing any way to use them in the early major-sequence and service courses that make up the bulk of our teaching load.

Now, there's a case to be made that in some ways the non-majors courses are even more important than the major-track courses--Miller cites general education science requirements in the US as a major factor in our relatively strong performance compared to other countries on his measures of public knowledge, for example. But I just don't get to do those classes that often, so I feel sort of cut off from this whole thread of academic development.

I know that there are people who do full-on inquiry-based education in the physics major track-- the Workshop Physics program at Davidson is the best-known example. This requires the entire department to buy into the concept, though, and while we've made minor moves in that direction at times, there's zero chance of implementing something like that in its full form.

It's really pretty frustrating. It's also hard to see how to make this work from a distance-- what I really need to do, I suppose, is to spend a semester or two teaching as a visitor in a department where they have a more inquiry-based program, and see how it works first-hand. But there are plenty of reasons why that's not going to happen any time soon, too.