By Joseph Groom
Cooper KM, Schinske JN, and Tanner KD. Reconsidering the Share of a Think-Pair-Share: Emerging Limitations, Alternatives, and Opportunities for Research. CBE Life Sciences Education (2021). Vol. 20: fe1. doi: 10.1187/cbe.20-08-0200
In their article in the March 2021 issue of CBE Life Sciences Education, K.M. Cooper et al. use recent studies to make a case for altering or abolishing the “share” portion of the widely used Think-Pair-Share method. The Think-Pair-Share is a popular active learning technique that allows students to come up with ideas on their own, bounce those ideas off of a classmate, hear a variety of student voices, and refine and articulate their own ideas by sharing them with the whole class. Cooper et al. succinctly explain how the “share” portion in particular can lead to inequities in learning, how certain assumptions about benefits of the “share” are not necessarily true, and how to effectively modify or remove the “share”.
How it challenged my assumptions and influenced my teaching philosophy
Before reading the article, I myself was making some of the assumptions the authors specifically identify. First, I thought that the “share” is the best way for the instructor to see how the students are understanding and processing the content in real time. Realistically, in the case of a voluntary “share”, most students who choose to share will reflect the perceived gender and racial identity of the instructor. Furthermore, men were found to dominate “whole-group shares” in an empirical study. The diversity of voices in a voluntary share will likely not reflect the richness or diversity of student ideas developed in the “think” and “pair” portions of the activity, in effect impeding the instructor’s appraisal of the class’s actual understanding.
Second, I thought that the diversity of voices being heard would increase with a random call version of the “share,” and with it, the classroom equity. While the diversity of student voices might increase, it turns out that students often spend so much extra mental energy on “sharing anxiety” that they cannot effectively focus on the learning outcomes. This is not limited to large classes. The “sharing” result can come out jumbled and confused, again impeding instructor’s appraisal of the student’s understanding. (But as a counterpoint, see this STEP blog post by Trisha Sippel about the benefits of random call.)
Third, I also thought that requiring students to talk in front of people is the best way to get them comfortable with public speaking. This also relies on random call for the “share.” When one does a simple calculation of how often a student will be asked to “share,” even in a class of 30 students, it is not enough to truly acclimate this student to the anxiety. Additionally, the authors point out that this form of scientific communication is very rare—normally, scientific professionals are asked to speak to a group about a topic they have considered for many years (rather than minutes).
Cooper et al. also posit some good modifications to the traditional “share,” in order to maintain the established benefits of understanding student thinking, honoring student voices, and gaining feedback for future instructional steps. My favorite modification would be “local sharing,” in which students share what they come up with in the “pair” portion to other groups or other rows, but not to the whole class. The “optional share” and a (seemingly) time-consuming “whip-around” are other alternatives, as is the “asynchronous share,” in which the instructor responds to and synthesizes written “shares” after class. While the “asynchronous share” would be the most equitable, it does seem to be a very time-consuming process for the instructor outside of class. However, it would be a good method of assessment, and could count towards the student grade if designed correctly.
A corollary for research seminars
Cooper et al. draw a very interesting parallel to the question-and-answer portion of scientific seminars, during which students or minoritized scientists reportedly “share” fewer questions and comments than certain established and powerful scientists. Overall, the authors provide compelling reasons to alter or even remove “sharing” from the Think-Pair-Share, and apply this logic to scientific presentations as a whole. I remember that during my first year of graduate school, a senior grad student I admired said: “I can seriously guess which questions all the bigwigs are going to ask after seminars.” They always got to ask them, though. At the time, I was grappling with the anxiety of asking questions at the seminar (or even formulating questions). I think a good practice to try would be to end the Q&A early to allow for small group or 1-on-1 questions at the front of the seminar room. Alternatively, give the entire group a few minutes to collect and formulate their questions. These modifications could encourage the shyer students to ask questions without experiencing “sharing anxiety.”