Do students learn any physics from doing labs? Often the answer is “no”.

Nobel Laureate in Physics and Science Education Crusader Carl Wieman recently weighed in on the value of laboratories in introductory-level lecture college physics courses on the discussion group of the American Physical Society’s Topical Group on Physics Education Research.  Our department chair, Horst Wahl, broadcast Wieman’s comments to the entire Physics Department faculty and it sparked an, uh, interesting discussion.  Here I share Wieman’s note and at the bottom attach the two papers that Wieman mentions in his note.  Enjoy.

In terms of research on learning from labs, there is very little research, but Natasha Holmes and I have been studying this recently. We did not look at studio format, but rather labs that are coordinated with weekly course schedule and done in different rooms and have the goal of supporting the learning of the course content. The attached PR-PER paper shows our results, which is that across 9 courses at 3 very different institutions, the labs consistently provide no measurable learning, with very small uncertainties.

Our attached paper on cognitive activities in undergrad research and the comparison to activities in intro labs explains why this result is not surprising. I also did a Physics Teacher article on this general issue of cognitive tasks involved in intro labs and how and why they are so different from what is intended. You can look up, was a couple of year or so.

It is very difficult to extract the value of the lab component to studio physics, because it is so completely integrated with all other aspects of the teaching. That would likely make the labs more beneficial. However, the data we got on the value of labs as used in somewhat different contexts, suggests one should exercise some caution in assuming they are of great value.

It would be a good research project to do a controlled experiment, comparing standard studio methods with an intervention where, in place of the labs students were given a simulation of the experiment, or the representative data from an experiment that was described to them, and measure the respective learning outcomes in the two treatments.

Carl Wieman




This entry was posted in Uncategorized. Bookmark the permalink.