Is Peer-led Better Than Instructor-led Active Learning?
When learning from textbooks, peer-led active learning led to higher exam scores, but for primary literature, instructor-led active learning was more effective.
Brasier’s Modern Biology course is an introduction to biology for both majors and non-majors that is designed to teach core principles of biological science and expose students to modern research. He employs a number of active learning strategies during class to augment students’ learning. To find out which strategies are most effective, Brasier investigated two different active learning strategies in his Modern Biology course of 91 students: instructor-led think-pair-share activities, and peer-led small group discussions led by former students trained as facilitators. All students experienced each of these activities twice, once during a class session covering a textbook chapter, and once during a session covering a research article. Readings did not differ in difficulty across conditions.
Brasier assessed student outcomes using targeted exam questions corresponding to the four different active learning sessions. For the textbook material, students performed better on exams when learning via the peer-led approach, compared to the instructor-led approach. However, when learning content from research articles, students performed better on exams after learning by instructor-led active learning strategies. These results suggest that both peer-led and instructor-led active learning strategies can have advantages, but their relative effectiveness may depend on the type of course content and/or reading material.
Figure1. Scores on exam questions were higher when students learned textbook content with an active learning activity that was peer-led (M = 76.74, SD = 9.89) versus instructor-led (M = 64.69, SD = 11.42). When the content was research-based, however, students performed better when the active learning was instructor-led (M = 94.95, SD = 6.66) vs. peer-led (M = 88.01, SD = 8.76). Error bars are the 95% confidence intervals for the means.
The mean difference of 12.05 for textbook-based content was significant, 95% CI [13.96, 10.14], t (90) = 12.52, p < .001, d = 1.31. The mean difference of 6.94 for the research-based content was also significant, 95% CI [8.85, 5.02], t (90) = 7.20, p < .001, d = .76.