Is "Think-Pair-Share" the Best Approach to Learning?

A traditional think-pair-share exercise (Lyman, 1981) might go something like this:

“Let’s do some learning, shall we?” I might ask my students. I then put on the board, or Zoom, or wherever, a question such as this: “What factors might make it more difficult to accept a client’s decision to terminate treatment with you? Take a minute or two to think about this and make some notes.” A minute or two later, I say, “Now, turn to the person next to you and compare your thoughts. Come up with some answers that are different from, and maybe even better than, what each of you came up with individually.”

As I move around the classroom (or the breakout rooms), I can see that all my students are talking, listening, sharing, and learning. At some point, I bring the class together and say something like this: “What did your pairs come up with?” Sometimes I’ll call on people randomly. This last question, the share part of the think-pair-share, is designed, in part, to (a) have students gain experience in talking in front of larger groups, (b) have the entire class benefit from other discussions, (c) have instructors get good information about what students are learning, and (d) solidify the learning that occurred in the thinks and pairs.

Pretty cool, right? In general, I, my colleagues, and the research say, “Yeah.” The think-pair-share is a basic building block of active learning, which is clearly an effective way to facilitate learning. But is the share a building block of the think-pair-share? In a recent article, Cooper, Schinske, and Tanner (2021) say no. They encourage us to re-think the share. I highly recommend reading their entire article. In fact, let me entice you by talking today about just a few of their objections and alternatives.

Problems with the "Share" Part of Think-Pair-Share

One major problem Cooper et al. see in the share is that the large-group discussion doesn’t bring out the best of the pairs. Thus, the share may not solidify knowledge as much as we think. They cite evidence that “the rich, equitable, and high-quality discussions seem to happen in small-group discussions, not whole-group discussions.” Indeed, it is possible that the students who volunteer to share may be the same ones who would contribute to (dominate?) a discussion even if there were no think and pair.

The share does not facilitate the goal of equity as well as we think. For example, Cooper et al. cite research to show that “whole-group discussions failed to capture the many high-quality comments made by women in small-group discussions.”

Another important point they make: Calling on students in the share may cause undue anxiety. Indeed, students who are thinking about having to present may learn less in their pairs. The anxiety some students fear may not be worth the increase in motivation that the share creates to be active in the think and pair.

Suggestions to Improve "Think-Pair-Share"

Cooper et al., being the educators and thoughtful people they are, are not content to articulate weaknesses—they present suggested improvements as well. For example, they suggest an “optional consent to share, whereby the instructor approaches individual students and asks if they would be willing to share their ideas with the class.” This approach is more respectful of student choice and retains the opportunity for students to practice presenting to larger groups. I certainly would have appreciated the opportunity to opt-out of some sharing when I was a student and having a particularly bad day.

Cooper et al. recommend a “local share,” including what I’ve seen referred to as “think-pair-square.” After the pairs discuss the question, they combine with other pairs, so students are presenting to three others, not just one and not the entire class. In my experience using the think-pair-square, I have noticed that students appreciate the opportunity to share their ideas with others without the anxiety engendered by "the whole class."

Cooper et al. present the option of instructors doing the share, which provides them opportunities to poll their students, summarize the best points from the pair discussions, correct misunderstandings, and shout out to individual students who made exceptional contributions to pair discussions. Instructors might also have students write their ideas on index cards (or discussion boards), and synthesize these ideas at a later time. Everybody gets more of a chance to think.

I really like Cooper et al.’s last suggestion: Just get rid of the share! “By eliminating the share portion… the instructor no longer dictates who gets to share… leaving more time for more students to get to talk with one another.” It may even leave enough time for another think-pair-sh—oops, I mean, think-pair!