PERbites

Prior experience with active learning methods promote adoption of such methods

Title: What really impacts the use of active learning in undergraduate STEM education? Results from a national survey of chemistry, mathematics, and physics instructors.

Authors: Naneh Apkarian, Charles Henderson, Marilyne Stains, Jeffrey Raker, Estrella Johnson, Melissa Dancy

First Author’s Institution: School of Mathematical and Statistical Sciences, Arizona State University, Tempe, Arizona, United States of America.

Journal: PLoS ONE 16(2): e0247544 (Open Access)


Students interested in pursuing a STEM degree start with an introductory course in an area of interest, say physics, mathematics or chemistry. Such gateway courses have a very important role to play in students continuing their studies in their field of choice. Unfortunately, such courses have high dropout rates, with negative learning experiences playing a very important role. Discipline Based Education Research has shown that active learning techniques (peer instruction, in-class collaborative problem solving, and so on) can lower dropout rates as well as improve learning in introductory classes. Despite this evidence the adoption rate of active learning by instructors is low.

In the paper we discuss here, the authors use a large scale survey to explore some of the factors that are frequently put forward as reasons for why there is a low level of adoption of active learning methods. The most interesting finding is that instructors are more likely to adopt active learning methods when they have prior exposure to active learning methods when they were students or have participated in education research projects.

The data

Education researchers have put forward suggestions as to why there is a lack of adoption of active learning methods including career concerns, lack of time due to research commitments, and so on. The authors decided to explore 6 of these common ideas: class size, classroom setup that promotes active learning, teaching evaluation from students and assessment of teaching, job security, research activities, and prior exposure to active learning methods. The authors chose these 6 since they are linked to policy decisions and priorities at educational institutions, and the results from such a study may help in guiding these policies and priorities.

The authors surveyed a large number of instructors from 2 year colleges, predominantly undergraduate 4 year colleges, and universities. The authors chose those instructors who were the primary instructors of general chemistry, single-variable calculus or an introductory quantitative physics course taught in 2017-18 or 2018-2019. Starting with a list of 18000 potential candidates, the authors filtered the list based on whether the instructors responded, whether the responses were complete and other factors, and ended up with a final list of 3641 participants.

In the survey, the participants reported the fraction of class time they spend in lecture mode, which the authors defined as students listening to the lecturer and engaging in typical problem solving exercises. The authors make the assumption that the time not spent in lecture mode is devoted to active learning techniques. In addition, the participants also answered questions related to departmental context, awareness of active learning methods and so on.

Results from the survey

The most interesting result from this survey is that, on average, instructors tend to adopt active learning methods when they have prior exposure to such methods either as students or as student instructors (or as a member of a team of instructors). In addition, instructors who engage in education research or funded education work are more likely to use active learning. Figure 1 (figure 6 in the paper) and the right panel of figure 2 (figure 5 in the paper), shown below, illustrate these findings.

Even though there is a lot of scatter, the average time used for lectures (as opposed to active learning) is lower by about 9% (statistically significant and small-to-medium effect) for those with prior experience with active learning than those without such experience. The difference is larger when we consider exposure to education research projects: about 12% which the authors report as being statistically significant and of medium effect (see right panel of figure 2 below).

Figure 1: Self reported percentage of time used for lecturing as a function of prior experience with active learning(figure 6 from the paper).
Figure 2: Self reported percentage of time used for lecturing as a function of research activity and exposure to education projects(figure 5 from the paper).

Another interesting finding from this study is that some of the ideas explored are consistent with commonly held beliefs. First, instructors in small class tend to use active learning more than those in large classes. Second, when classes are designed to enable active learning the adoption of active learning increases. Moreover, when assessment of teaching is important then use of active learning increases. On the other hand, when student teaching evaluation is important the use of active learning drops.

Interestingly the study also finds that contrary to what we might expect, the time that an instructor spends on research activity doesn’t seem to have much effect on the use of active learning. Finally, the study also finds that the common perception that instructors with job security might use more active learning than those without job security is not consistent with the data. Both of these groups seem to use active learning at similar levels.

The authors report that the results represent nation wide trends but, as the spread in the data hints at, there are many instructors who are exceptions to the general national trends. For example, there are some instructors who spend most of the time on active learning methods, and some who use rarely use such methods. In future research the authors intend to look at such trends and try to understand why the instructors who are exceptions to the national trends are so.

As with all studies, this study also has several limitations. The most important limitation is that the data are self-reported by the instructors and are not observations of what instructors actual do . Moreover, the results presented here are correlations and hence we can’t make casual inferences. Finally, the study doesn’t compare more than two variables at the same time. This means that we may be loosing information present in interactions between the variables.

Implications of the study

The main takeaway from this study is that prior experience of instructors with active learning methods or their experience working in education research projects drive adoption of active learning methods in class rooms. This implies that enabling students to participate in classes involving active learning methods may increase the likelihood that when these students become instructors they will adopt such methods. In addition, enabling instructors to receive local and national professional development opportunities may lead to similar outcomes. Finally, providing instructors with smaller classes as well as class rooms that are designed to encourage active learning may also increase the adoption of active learning methods.

Figures used under Creative Commons Attribution 4.0 International. Header image used under Attribution-NonCommercial-ShareAlike 2.0 Generic (CC BY-NC-SA 2.0) from Flickr user MKSeery.
Exit mobile version