Authors: Miguel Rodriguez and Geoff Potvin
First author’s institution: Florida International University
Journal: Physical Review Physics Education Research, 17, 20131 (2021) [open access]
By now, it’s nearly impossible to be in higher education and not have heard about active learning or seen active learning advertised as a solution to nearly any student learning problem. After all, research has shown that active learning does increase learning and can reduce failure rates.
Yet, active learning is a broad term that doesn’t give too much description of what specifically the instructor is doing other than not lecturing for the entire class. Therefore, if an instructor wants to implement active learning based on a research study, they need to figure out what the researchers actually did.
Further, when results are reported broadly, it can be difficult for instructors to know what parts of the active learning implementation were responsible for better student learning. For instructors with limited time and resources, they may be forced to only implement parts of the intervention.
The goal of today’s paper is to focus on active learning strategies and see how they affect student learning, finding that working in small groups each class period seems to be one of the most effective.
To reach their conclusion, the researchers used responses to the Conceptual Understanding and Physics Identity Development Survey, which measures students’ career intentions, previous science learning, prior academic performance, and details about physics courses they’ve taken in a pre/post format. They sent the survey to a representative sample of physics departments across the United States to cover a variety of institution types and include students in both algebra-based and calculus-based introductory physics courses.
In addition, to measure student learning, the authors also asked students to take a portion of the Force and Motion Conceptual Evaluation (FMCE). To reduce the length of time it would take students to complete all research activities, the authors only asked students to answer the 17 questions that showed the strongest validity and reliability based on prior research.
After matching pre-course responses with post-course survey responses, the authors had data from 371 students from 19 unique institutions. They then used linear regression to describe FMCE scores based on various active learning practices students described in their physics courses in the survey.
When looking at the results, the researchers found that students who reported working in groups every class scored significantly higher than students who did not report working in groups every class. When comparing students who worked together in small groups every class period to students who never reported working with other students in groups, students in the first group answered 2.6 more of the 17 questions correctly than students in the latter group.
Other practices such as using lab equipment, working with other groups, using computer simulations, and working on labs or projects everyday in class were also found to result in higher FMCE sources, though to a smaller degree than working in groups every day.
While using these practices once a week in class occasionally resulted in an increase in FMCE scores, the results showed no consistent pattern outside of using a practice once or twice a semester. In that case, using an active learning practice once or twice a semester never resulted in better FMCE scores compared to never using the active learning practice.
Next, the researchers wanted to learn more about what the learning environment looked like for students who worked in groups every class. Using the results of the survey, the researchers found that students who worked in groups every class were more likely to work at tables, face their peers, use lab equipment at least once a week, collaborate with other groups, simultaneously use numbers, formulas, graphs, and words to solve physics problems, use computer simulations, and work on labs/projects than students who did not work in small groups each class. Other common features of active learning classes like using whiteboards or presenting work to others did not differ in reported frequency between students who worked in groups each class and those who did not. (Figure 1).
When trying to explain why there might be such a benefit to working in small groups, the researchers thought there could be a variety of explanations. First, students might not want to look bad in front of their classmates and hence, try harder when working in groups. Students might also feel a greater shared responsibility to their teammates than if they were working alone.
Second, working in small groups regularly might have allowed students to learn from each other and receive more immediate help than would have been possible in a large lecture course.
Finally, regularly working in groups might have helped students with social integration and forming connections outside of class with other students. Prior work has found that students with broader networks tended to perform better than their peers with smaller networks.
Taken together, the results suggest that instructors interested in adding active learning practices to their courses might want to start with small group work. However, for a noticeable improvement in student learning, students need to work in these small groups every class period. The benefit from working in small groups once a week was not statistically different than never working in small groups during the class!
Furthermore, using other techniques in each class like using equipment, working with other groups, using computer simulations, and working on labs and projects seemed to be effective too, though to a lesser degree than regular group work.
Despite these findings, the use of such techniques in introductory physics courses is limited, based both on prior work and the fraction of students who reported their physics course using these techniques. The authors then suggest that future work should look into how to achieve widespread adoption of effective teaching practices in physics.
Figures used under CC BY 4.0.
I am a postdoc in education data science at the University of Michigan and the founder of PERbites. I’m interested in applying data science techniques to analyze educational datasets and improve higher education for all students