Are active learning pedagogies really a "big deal"?

Martella et al (2023) call into question the empirical research on active learning pedagogies.

Active learning pedagogies are “kind of a big deal” nowadays.

The idea is straightforward: lots of research evidence shows that people aren’t blank slates upon which lessons are written, but rather they construct knowledge by integrating new ideas with ideas they already have. Therefore, it seems reasonable to design instruction to foster students’ active knowledge construction (e.g., solving problems, group discussion) rather than passive receipt of information (e.g., lecture). Of course, things aren’t quite that simple.

Even the biggest proponents of active learning pedagogies acknowledge there’s a time and place for lecture (e.g., when most of the students in a class are confused about an idea, a quick lecture can be really useful). But, nonetheless, there’s been a big movement to push instructors away from the “sage on the stage” model (i.e., lecturing all class, every class) to including more active and interactive activities in their pedagogy. Given how entrenched lecture has been in education, advocates have had to push hard to get instructors to try different, more active pedagogies.

Much of the enthusiasm around active learning stems from a meta-analysis done by Freeman et al. (2014), which made a strong case that active learning pedagogies were better for students than lecture. But, a new article by Martella et al. (2023) suggests we should tone down our confidence in Freeman et al.’s findings, as well as findings from subsequent studies of active learning pedagogies. Martella et al. coded the studies in the Freeman et al. meta-analysis, and a random sample of active learning studies published from 2015-2022, for threats to internal validity. They found numerous threats in both the Freeman et al. corpus and in more recent studies. Yikes!

Now, Martella et al. acknowledge that it is difficult to do authentic, in-classroom research and control for every threat to internal validity. Often, reasonable concessions must be made. But, the more threats, the less confident we should be in the findings. And, across all the studies they coded, all of them had at least one internal validity threat, with the majority having six or more. Yikes yikes! So, does that mean we should all go back to lecturing all the time, every time?

I don’t think so, and I don’t think Martella et al. do, either. Instead, I view their work as a great guide for designing future research (and grant proposals!) on active learning pedagogies. Researchers should do everything they can to eliminate threats to internal validity, and when they cannot, they should analyze their data in ways that account for them, as much as possible. And in every case, the authors should provide enough detail so that readers can decide whether the threats to internal validity really do invalidate the findings or not. Those are great lessons for researchers of active learning pedagogies but also all education researchers. So, the article is definitely worth a read, for those lessons alone.

What do I think about the active learning literature after having read this article? Well, like most research done in real-life settings, it’s imperfect. But that doesn’t mean it’s directionally incorrect. Martella et al. recommended more research on specific pedagogical strategies to promote cognitive engagement and also more research on how to integrate lecture with participatory class activities - that makes sense to me. And until that research is done, I’m going to continue trying to find the best tool for every learning objective in my courses - and I suspect some of those tools will be active learning pedagogies, and others will be lectures.