A Paean to Think-Pair-Share

I had originally thought to title this essay “An Ode to Think-Pair-Share.” But I decided that rather than simply a poem written on a particular subject, I wanted to write a text that expressed the enthusiastic praise for my personal pedagogical panacea.

Something like the hot whiskey sling my grandmother used to give my father for each and every childhood ailment he had, the pedagogical technique known as think-pair-share (TPS) is my equivalent of a cure for what ails my classroom: Is attention flagging? TPS! Is Section Kid dominating the conversation? TPS! Want to encourage active learning and the social construction of knowledge? TPS! In this essay I’ll share some of the ways I use think-pair-share.

First, a brief description of the process: you pose a question and have students think about it for a minute or two, perhaps jotting down a few possible responses in their notebooks. This is pedagogically sound, because everyone is engaged and active; everyone also has something to say. Then the students pair up -- you could even put them in groups of three or four -- and discuss their answers, trying to come to a consensus. While they do this, you can go from group to group, either listening to get an idea of what they are thinking or even nudging them toward answers. Finally, someone from each group must share one or more of the group’s answers.

Think-pair-share is so useful because it’s a general-purpose didactic instrument, a veritable Swiss Army knife of classroom tools. TPS is simple: you, the instructor, need no materials at all. It requires zero prep work other than coming up with a question.

What’s more, TPS is an active-learning device: it forces everyone in the class to think and to write something down. During the pair phase, students can compare notes, something that a great deal of research shows increases not only comprehension but also retention of the material.

Finally, the share phase, if done correctly, means that you can better modulate the discussion to involve everyone. Rather than just posing a question for the group and hoping that rather shy student raises their hand, TPS allows you to call on that student and ask for their group’s consensus.

Some Tips for TPS

TPS’s simplicity belies its potential for nonoptimal use. For the first step, don’t just have the students ponder something and then pair up. Ask them to write something down in their notebooks, and let them know -- especially the first time you deploy the technique -- that they’re going to compare with their neighbors. Make sure you both say and have visible the written prompt.

Another misstep is always using the same group or groups that are too large. One time, you can have students work with three other nearby students, but if one of your goals is for students to get to know each other, have them count off and then have all the ones, all the twos and so forth group up. Four people is, I think, the optimal size for these groups. You should direct students to introduce themselves to the rest of the group before they start the group deliberations.

While the groups are working, you should circulate around the room. I’ve found that if I squat next to the groups, they’ll include me in the conversation, whereas if I simply stand next to them, I can eavesdrop without them addressing me. This is an opportunity to either nudge the group in one direction with a question to them or to simply assess whether the groups are all heading in the same direction.

The share part of the exercise is the one most fraught with mistakes when I see my teaching assistants executing an otherwise perfect TPS. First, you, as the instructor, should direct most of the sharing. If you simply ask for volunteers, you annul a fabulous feature of the TPS: getting some students to talk and getting other students to shut up and leave some space for quieter classmates. The barriers are different: some students are nervous about raising their hand and speaking aloud an answer that might be wrong. Other students don’t realize that they have spoken far too much already.

It is therefore better if, at least initially, you decide who will report. That said, it’s absolutely crucial to make explicit, every time you call on someone, that they are reporting for their group. If the answer is wrong, it’s only partially that person’s response. I always say something like, “Sarah, what did your group come up with?”

For what it’s worth, I almost always use TPS to ask open-ended questions with lots of answers -- for example, “What are some of the reasons American foodways changed during the 19th century and what are some primary sources we could use to show that change?” -- rather than questions that have only one correct answer. All of this lowers student anxiety and makes reporting to “warm-calling” (as opposed to cold-calling) a core expectation of the class.

Think-pair-share works great in a seminar-size class, but I’ve also used it in courses with 100-plus students, both in person and over Zoom. For large in-person lectures, I had students sit in their 15-person discussion sections. I would pause my lecture and ask the students to discuss with the three people nearest to them, then solicit answers from a few people. It’s not possible to do the same sort of random assortment, but the pedagogical benefits are the same. It’s worth making explicit to the students that the questions will always be open-ended and that speaking in front of others is a good skill to practice in a low-stakes environment.

On Zoom, you’ll face other challenges, but TPS is still an option -- though only up to around 100 students. When I had 98 students in fall 2020, I could put them in breakout rooms of four people. The thinking and pairing went on smoothly in breakout rooms, and the sharing happened when I called on students after the breakout rooms closed. This past spring, with 237 students, five or even six students were in each room (because of Zoom’s limitations on number of breakout rooms). The higher number of people in each breakout room led to many students reporting that their peers immediately turned off their cameras and refused to participate once in the breakout room. I tried random TA “pop-ins” to the breakout rooms, but even that didn’t help compliance. Students reported that it was less anxiety-inducing if we told them ahead of time who was going to report -- say, the person who woke up earliest that day, the youngest person in the room or some such -- before simply calling on them randomly. Some initial research suggests that despite the difficulties of pulling off TPS remotely, it does improve student performance.

Think-pair-share has become my go-to for creating an active, equitable sharing of ideas in the classroom, whether physical or virtual. That said, it’s also useful for other purposes: my head teaching assistant suggested we use it at the beginning of every instructor meeting to get an idea of how the TAs’ discussion sections had gone and the common problems they were seeing on student papers. I’ve also engaged in TPS with colleagues as we worked to draft changes in our curriculum. It’s a great first technique to try what James Lang calls “small teaching” -- a seemingly minor change or new practice that can reap significant educational rewards.