Improved Learning in a Large-Enrollment Physics Class (2011) [pdf]
bu.edu
bu.edu
1) The proponents seem to ignore a ton of research in education dating back to the 80s, or cherry pick certain facts to support their desired interventions. A good example of this is the research program that sprang up around cognitive load theory. Wieman (the last author on this paper, and a Nobel prize winner in physics) talks a lot about how working memory constraints should shape the way we teach students physics, and cites certain research related to working memory (for example [0]). But a lot of that research indicates that some of the strategies he and his institute suggest might be ineffective [1].
2) Both the linked paper and another well-known work [2] lump together a large number of heterogeneous interventions under the heading of "active learning" and conclude that "active learning" "works." But it's not clear which particular interventions are effective, and which are ineffective or perhaps harmful. In fact, this information is impossible to tease out from these "shotgun" style studies. So to what extent should I take any particular intervention that Wieman recommends seriously?
3) I'm not convinced these studies ever appropriately control for experimenter/teacher effects.
This is not to say that I don't think there are important lessons college instructors can take from educational psychology. Just that his current strand of "active learning" research (coming from Wieman and others) should be taken with a grain of salt.
[0] https://cwsei.ubc.ca/sites/default/files/cwsei/about/Wieman-...
[1] https://www.tandfonline.com/doi/abs/10.1207/s15326985ep4102_...
Long version (about math, but with specific, non-trivial recommendations that obviously generalize to some extent): https://www.amazon.com/How-Wish-Taught-Maths-conversations/d...
Another HN commenter suggested the book to me recently. It's really good, and gives copious citations. Some gripes: It's way too long, and there are a handful of things I feel are better explained elsewhere (e.g. the "good" example used when talking about how to label diagrams I feel is actually bad...).
I’ve never seen clickers used effectively and they were used heavily in eight years of my education. That is, I’ve never seen them used in a way that could not have had its results replicated without the use of clickers.