Think, Pair, Share was popularized by Lyman (1981). He gave it visibility and although it is now four decades since he described it, relatively few instructors seem to know about it, or know how and when to use the methods. Note, other teaching innovations are also long in the tooth. For example: Mazur (1991) peer instruction three decades, Robbins and Folts (1932) case based teaching at least eight decades, and perhaps we should stop describing such common sense, long standing ideas as innovation and just call them best practice; period, end of discussion. Having students work together was not new in 1981. Working with real world examples to study was not new in 1932. These practices make sense!

It is like discussing the benefits of having learning outcomes for a course, and sharing them with students at the beginning of the course, and convincing an instructor to use and share learning outcomes; do we really need to keep talking about this? Really? Why are we still having these discussions?

In the think-pair-share method, the basic idea is to have the student think about something on their own first, then pair up with another student to discuss what each other did, and then share their final thinking and results with the class. This can be done in assigned homework activities, labs, tutorials, or in the primary lecture time. This is good common sense to do. It makes sense. Unfortunately, the devil is in the details.

Observations…

• What the students are given to think about needs to be carefully thought through with respect to the student’s ability to do self-thinking, self-learning, and the assumptions about existing knowledge and skills
• The activity needs to be slightly structured with suggested or regulated timing for each of the tasks
• Combinations of the think-pair-share activities can be used for reinforcement and deliberate practice, within the same lecture.
• The activities can be flipped with the students exploring something yet untaught, and then the instructor in the debriefing and follow-on lecturing delve into the science and foundations.
• With today’s learning environments, the students can upload their individual thinking and the paired result; the paired summary can explain the source of the various ideas, what was accepted, what was duplicated, what was dropped.
• The activities can be cascaded where you start off single then with the pairing of students, then with four students, then with eight students, then 16 students; where at each iteration students are learning about sharing ideas, vetting, and rationalization.
• The scaling up works well with meaty topics, and it is important to continue the consolidation process and documenting the rationalization that provides a provenance of sorts for the ideas and for each student’s contribution.
• Even though each part may not be marked or assessed in terms of a rubric or feedback, the uploading of the solo, paired documents can be used to identify students at risk, different levels of mastery and contribution, which can be used in a final assessment for the term.
• If you have these activities every lecture or every few lectures, there will be lots of data points for you to review if you need to think about a student’s contribution, engagement.
• While think-pair-share activities should not be used for attendance marks (we do not believe in just for showing up), or actually marked per se, the think-pair-share activities do give you some idea about the students’ engagement and their activity level during the term.

• The first time that the students do think-pair-share, instructions should be given about the intent and approach to be used, and this debriefed after the first activity to make the students self aware and mindful of how they are doing the think pair share as well as the actual result of the activity.
• With larger classes, it is not likely that all of the pairs will be able to share with the class and it is important to rotate through the pairs if possible.
• There can be the tendency for certain students to over-volunteer with a chance of dominating the class, so it might be best to number the pairs or number the groups and randomly pick numbers to present.
• If it is not possible to have everyone share in class time, the think-pair-share sessions can be reviewed by a TA or the instructor after class, and a general discussion, commentary can be made in the class announcements or in the next lecture.j
• It is possibly best, if the pairs are held stable for the first part of the term as this will reduce uncertainty and create some stability in the process with the students learning to work with each other, and then later in the term the pairs and groups can be switched up and changed.
• It is suggested that the pairing, is random if possible, without the students working with their best friends and study buddies; if the students work with someone they are comfortable with and who they study with all of the time, the diversity and different ways of looking at a problem will be minimized.

Further reading

• https://dictionary.apa.org/social-comparison-theory
• Robbins, E.C. and F.E. Folts (1932), Industrial Management – A Case Book, McGraw-Hill.
• Lyman, F. (1981). The responsive classroom discussions: the inclusion of all students. A. Anderson (Ed.), Mainstreaming Digest, College Park: University of Maryland Press, pp. 109-113.
• CTE Teaching Tip: Peer Instruction and Concept Tests
• Kahneman, D. (2011) Thinking, Fast and Slow, Farrar, Straus and Giroux.
• Kahneman, D., Sibony, O. and C.R. Sunstein (2021) Noise: A Flaw in Human Judgment, Little, Brown, Spark.