Active learning works.
Some would say the results were already clear. Active learning , in which students spend class time actively engaging with material, by thinking about it, practicing skills, or discussing it with their peers, should result in better learning than the traditional instruction model of passively sitting in a lecture hall. After all, in order to improve, one needs practice and feedback, both of which active learning provide.
For those that remain unconvinced, a recent metastudy adds to the already substantial set of evidence in favour of active techniques.
In this work, appropriately titled “Active learning increases student performance in science, engineering, and mathematics,” Freeman et al. metaanalyze 225 studies comparing active learning to traditional lecturing in science, technology, engineering, and mathematics (STEM) courses . The main findings of the study are twofold:
- Active learning reduced the failure rate, compared to traditional lecturing, from 34% to 22%.
- Active learning increased the performance on test scores, compared to traditional lecturing, by nearly half of a standard deviation (an effect size of 0.47).
If one restricts to the studies which compared performance on concept inventories, the effect size increases to 0.88, or almost a full standard deviation. This is significant: Most concept inventories are validated and reliable assessments of higher level cognitive skills, in contrast to course exams which often focus on lower level goals, such as simple recall and ‘plug-and-chug’ quantitative problems. This means that active learning has a greater impact on the type of high level learning that many university STEM courses aim for.
Focussing on the aggregate results, the effect sizes reported are large. The authors note that an effect size of 0.20 or smaller for K-12 education is interesting from a policy perspective; here, active learning produced an average effect size of 0.47 for performance increase. A comparison for the effect on failure rate is more alarming: The authors point out that if this were a controlled medical trial, the experiment may have been discontinued because the treatment (in this case, active learning) was clearly more beneficial than the control (lecturing).
In his commentary on this article, Carl Wieman summarizes this discrepancy :
Although more effective teaching methods have been overwhelmingly demonstrated, most STEM courses are still taught by lectures—the pedagogical equivalent of bloodletting.
There are many factors affecting the decision of faculty to adopt active learning methods, both personal and institutional. Much work remains to be done to facilitate system-wide reform towards active learning. At the least, it is now overwhelmingly clear that the need for this reform is firmly based in evidence.
 See this newsletter from the Carl Wieman Science Education Initiative (CWSEI) group in the Department of Physics and Astronomy at UBC for examples of active learning techniques.
 Freeman, Scott, et al. “Active learning increases student performance in science, engineering, and mathematics.” Proceedings of the National Academy of Sciences (2014), 111:8410-8415.
 Wieman, Carl E. “Large-scale comparison of science teaching methods sends clear message.” Proceedings of the National Academy of Sciences (2014), 111:8319-8320.
Footnote: See also this related post by the Life Sciences division of the CWSEI at UBC.