student lab screencasts

At the beginning of this past semester of teaching Physical Optics, I used this blog to help me think about some ideas of how to incorporate labs into my standards-based grading scheme (here’s part 2). Now that the semester is over, I thought I’d get down some thoughts about how I thought that worked.

First of all, I didn’t want to lose the very important idea of teaching students how to write formal lab writeups. I thought I took care of that with “I can write an organized lab report” as one of my standards. I told the students that they could pick any lab they wanted for that standards, but that they would get essentially two scores for it: one for the lab standard (like “I can measure the speed of light”), and one for the formal report standard. One motivation for that came from past years with the formal nature of the reports (for every week, back then) really weighed heavily on students and they fell behind. Unfortunately, I’m here to report that this standard did not do what I was hoping it would. No one turned anything in on this until the last two weeks of the semester and few of them got good scores for them because they didn’t allow enough of a feedback cycle. I didn’t have a “two week” rule for that standard, but now I see that I should have. What that would mean is that a students would have to get a first draft of the formal report in within two weeks of whenever that lab was done.

Student screencasts

For the rest of the labs, students needed to give me a screencast walking me through their setup, theory, data, analysis, and conclusions for a given lab. Often this turned out to be a series of screencasts, but that was ok by me.

Lab screencasts shared many positive qualities with the normal screencasts students were doing for the rest of the material. I felt that I could really get a handle on how well they understood things, mostly at the set-up and analysis stage of things. The students almost always had partners for the labs, but still had to do their own screencasts. Often I would see the same diagram for a setup multiple times, but with slightly different (and sometimes wrong) descriptions of it. I liked that a lot better than just getting one formal writeup from everyone in the group, or even getting a different writeup from everyone in the group, since then there was often similar language. Hearing a student describe something, even something that is written or drawn correctly, is a really eye-opening experience.

My hope with the screencasts was that they would feel like less of a burden than the formal writeups I used to assign (every week). While I think that was true, I still got students feeling like they didn’t know exactly how to go about the screencasts. When they would articulate those thoughts, I’d always say that they could either come talk to me about it, or just get whatever they understood about the lab into a screencast, to give me a chance to give them some feedback. They rarely did the latter, but it’s something I want to work on in the future, essentially to find a way to lower the psychic barrier between them and getting a “rough draft” screencast in.

I really liked how the screencasts aided in my assessment of their analysis skills. It’s much easier for me to see how well they understood the analysis when I can watch them doing it, typically in Mathematica. In the old days, I’d either get a Mathematica notebook turned in that (often) wouldn’t work for me, or I’d get just the results pasted into the formal report. Once again, if they leaned too heavily on their lab partners for the analysis, I could tell by listening to them.

The biggest downside to the screencasts is the time I spent grading them. In this case, though, I would say that extra time almost always meant a much better understanding of how well they knew things, and much better feedback that I could give them. What I (mostly) mean about that, is that in the old days I would look at a stack of lab reports and scan them quite quickly. Because I couldn’t easily figure out who had done what, I would give them the benefit of the doubt and they’d all get a decent score. I know there’s ways to do that better, but I think I like this screencast approach, even with the extra time, than the methods I’ve learned about to help assessing group written work.

My friend Matt Vonk has done some different work with students producing video for lab work. 

Anyone else?