These quick tips from the Faculty Teaching Institute can be used with think-pair-share (TPS), where students think on their own, pair to talk to a neighbor, and share with the whole class, with or without polling.

Explain to students why you are using think-pair-share. Explain that the point is to help them learn by reasoning and discussion. Otherwise, they may assume that the goal is to test or track them.

Offer credit for participation. This may be based on observations of student behavior, and/or points through electronic clickers (e.g., 1 point for answering, extra credit for the correct answer).

Use a variety of questions, including conceptual questions. Many existing TPS questions focus on conceptual understanding, which is useful for student learning. You can also use TPS questions for surveys, interpreting graphs, learning reflections, predicting outcomes of demonstrations, or breaking down a problem into parts.

Use a mix of simple and complex questions. Use a range of simple and complex questions to help students build confidence, and also engage with difficult ideas. Many people use questions that are too simple. If a student can answer the question on their own, then it’s not a TPS question.

Give students time to think about their answers on their own. Let students read through the question and reason about it to arrive at their own answer; this avoids privileging students who are faster readers or thinkers. Ensure you give them enough time to do this by reading the question and answering it on your own. Then ask, “Does anyone need more time?” (not “Is everybody ready?”).

Have students vote anonymously (for TPS with polling). If you are using ABCD voting cards, have students vote on the count of 3, and hold cards against their chest so other students can’t see their votes. They can use the back side of the sheet for “none of the above” and all four colored squares for “I don’t know.” With electronic polling, do not show students the results of this initial vote.

Decide whether to have students turn to their neighbors (for TPS with polling). A common rule of thumb is that students may not need to discuss their answers if 80% or more vote correctly.

Circulate and listen in while students discuss. This is very valuable and lets you listen to their reasoning, and identify some ideas that can be brought out in the large group discussion.

Have a signal for ending small group discussion E.g., a bell, raising your hand, or telling “time!”. Give them a 10-second warning. Then (for TPS with polling), have them re-vote.

Wait to reveal the answer. If you reveal the correct answer as soon as students are done voting, it can short-circuit students' engagement and thinking and convey that the right answer is the only thing that matters. Hear reasons for all answer choices before confirming the correct answer.

Hold an interactive debrief, discussing the reasoning behind right and wrong answers. Whether you have students discuss with one another or not, the debrief should always include a discussion of what the correct answer is, and why the incorrect answers are not correct. If no students volunteer to explain a particular answer, you might suggest some reasons. Note: If students have not yet converged on the correct answer, there may be something confusing about the question.

Tips for writing TPS polling questions:

• Use existing questions. See lists of question banks at https://cwsei.ubc.ca/resources/instructor/prs or in Peer Instruction (Mazur, 1997).
• Draw from your students for answer sources. Some good sources of TPS questions are the questions students ask you, or common analogies you have found that help students.
• Don’t make them too easy. This is a common mistake and misleads students as to your expectations. Use questions that emphasize reasoning and challenge student thinking.
• Include tempting distractors. Use distractors (the incorrect answer choices) that represent likely student answers or difficulties, so that students will actually vote for them. A good technique is to use students’ wrong answers to open-ended questions as your distractors.
• Get feedback from colleagues. Or just try the question with students and revise next time.
• Write or find questions that target student difficulties. Identify an area that students struggle with, and use a question to help them wrestle with the idea. The best questions are those which only 30-70% of students answer correctly before discussion. Online sources have great questions you can use; past exams and homework are also good sources.

Additional practices:

• Use questions to set up instruction, not just assess learning. While TPS is great as an exit poll and to keep students on target, it can also be used to assess prior knowledge, provoke thinking, and elicit misconceptions. You can use TPS questions to lead into lectures too: Asking students to predict the outcome of a surprising demonstration can then create a “time for telling.”
• Just show students the question without a lengthy introduction. Give students a prompt such as “I’ve got a question for you.” Don’t go into a lengthy introduction of the question, and avoid reading the question out loud. This allows students time to read and interpret the question.
• Tell students to discuss with a phrase that tells them what you want them to do. Some good phrases are “Turn to your neighbor and convince them you’re right. Just because you both have the same answer does not mean that you're both correct. So make sure you explain your reasoning. You've got about a minute. Go!”

This Expert Recommendation is based in part on G. Brissenden & E. Prather, Think-Pair-Share: A revised “How to” guide (2015), D. Duncan, Tips for successful ‘clicker’ use (2008), and D. Bruff, Teaching with classroom response systems (Jossey-Bass, San Francisco, 2009).

This material is based upon work supported by the National Science Foundation under grants DUE-2141678, 2141745, 2141769, 2141795, and 2142045.  Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation.