Peer Instruction¶

Strategy · All strategies
Evidence · 1 claim (1 for) · 3 studies (2 quant-synthesis, 1 theoretical), q2–q4 · 2 of 3 report an effect size

Description¶

Peer Instruction, developed by Eric Mazur at Harvard, structures class time around a repeating cycle: the instructor poses a conceptually demanding multiple-choice or true/false question (ConcepTest); students commit to an individual answer, often via clicker or show of hands; they then attempt to convince a neighbor with a different answer; finally they revote and the instructor debriefs based on the vote distribution. The strategy converts lecture time into episodes of active learning driven by cognitive conflict and peer explanation.

Design Implications¶

Peer Instruction operationalizes the finding that interactive engagement outperforms passive lecture on conceptual measures Interactive engagement methods outperform traditional lecture on conceptual understanding. [+S]. Its engine is the discrepancy between a student's initial answer and a peer's reasoning: articulating an explanation and confronting contradiction forces retrieval and elaboration that listening alone does not produce. Questions must target conceptual understanding rather than calculation — students should be able to argue about them without paper-and-pencil work.

Context¶

Requirements¶

• Conceptually focused questions with plausible distractors that expose common misconceptions
• A mechanism for individual commitment before discussion (clickers, cards, or online polling) to prevent anchoring on a confident peer's answer
• Sufficient class time for the full cycle (typically 5–10 minutes per question)
• Instructor willingness to adapt the debrief to the actual vote distribution rather than a planned script

Constraints¶

• Fails when questions are procedural or rote — there is nothing substantive to argue about [~M]
• Students with low confidence may defer to peers without genuine evaluation, propagating misconceptions when few students hold the correct answer [~M]
• Requires a critical mass of students who have done preparatory work; in an unflipped course, students meet the question cold and discussion quality drops
• Large vote majorities (>90% correct) make discussion pointless; small majorities (<35%) may leave no student able to articulate the correct reasoning

Implementation Variability¶

• Flipped Classroom variant: content delivery moves to pre-class reading with Just-in-Time Teaching quizzes, freeing class time entirely for Peer Instruction cycles
• Two-stage or two-round exams: the same discussion mechanism applied to assessment, where students first answer individually then discuss and re-answer
• Online adaptation: asynchronous discussion boards or breakout rooms in video conferencing replace shoulder-to-neighbor talk, at some cost to immediacy

Target Learners¶

• Undergraduate and advanced secondary students in conceptually dense domains (physics, chemistry, economics, biology) Interactive engagement methods outperform traditional lecture on conceptual understanding. [+S]
• Students holding entrenched misconceptions, which surface and become arguable through the voting cycle
• Less effective for complete novices who lack enough prior knowledge to generate or evaluate any explanation — pair with Activation of prerequisite knowledge first

Target Learning Goals¶

• Conceptual understanding and misconception repair, rather than procedural fluency
• Argumentation and explanation skills: justifying reasoning and critiquing others'
• Metacognitive awareness of one's own understanding, exposed by the gap between confidence and correctness

Instructions¶

1. Assign preparatory material and optionally a pre-class quiz so students arrive with baseline exposure (Flipped Classroom).
2. Pose a ConcepTest targeting a known misconception; students vote individually without discussion.
3. Display the vote distribution. If answers are split (roughly 35–90% correct), have students discuss with a neighbor holding a different answer.
4. Students revote. If the majority now answers correctly, briefly confirm and consolidate; if not, reteach or run a second round (Class Discussion).
5. Repeat with 2–4 questions per class session, keeping each cycle under ten minutes.

Related Strategies¶

• Think-Pair-Share — the same individual-then-discuss structure without the voting and revote mechanism
• Just-in-Time Teaching — supplies the pre-class preparation and question targeting that Peer Instruction depends on
• Clicker Questions — the polling infrastructure that enforces individual commitment before discussion
• Peer Instruction structured questioning sequence: individual answers, convince-your-neighbors discussion, revote, and instructor feedback

Examples¶

• Peer Instruction Network / Mazur Group resources — Eric Mazur's original implementation in Harvard Physics 1, replacing most lecture with ConcepTests; documented in Peer Instruction: A User's Manual (1997).
• Carl Wieman Science Education Initiative — Clicker resource guides and question banks from UBC and CU Boulder, where large-enrollment science courses adopted Peer Instruction at scale.
• Perusall — extends the peer-discussion mechanism to pre-class reading, with students annotating and responding to each other before the vote cycle begins.

Key Sources¶

• Mazur, E. (1997). Peer instruction: A user's manual. Prentice Hall.
• Crouch, C. H., & Mazur, E. (2001). Peer instruction: Ten years of experience and results. American Journal of Physics, 69(9), 970–977. doi:10.1119/1.1374249
• Freeman, S., Eddy, S. L., McDonough, M., Smith, M. K., Okoroafor, N., Jordt, H., & Wenderoth, M. P. (2014). Active learning increases student performance in science, engineering, and mathematics. PNAS, 111(23), 8410–8415. doi:10.1073/pnas.1319030111
• Smith, M. K., Wood, W. B., Adams, W. K., Wieman, C., Knight, J. K., Guild, N., & Su, T. T. (2009). Why peer discussion improves student performance on in-class concept questions. Science, 323(5910), 122–124. doi:10.1126/science.1165919