Peer instruction: join individual answers with discussion and a second check

The key is explanation, not polling

Peer Instruction asks students to answer a conceptual question individually, explain their reasoning to a partner or group, and answer again. Polls can use paper, hands, or clickers; record the first response before discussion (Crouch and Mazur’s classroom study).

A hypothetical lesson and assessment

Suppose a physics class asks, “When a heavy truck and a small car collide, are the forces they exert on each other equal in size?” Each student chooses and writes one reason alone. Partners explain where their reasoning may confuse Newton’s third law, then vote again. The instructor clarifies the paired forces and gives a new individual question with a different context. This is a lesson example, not measured evidence.

The 35–70% first-vote range comes from Crouch and Mazur’s particular introductory-physics setting, where they observed discussion gains. It is not a universal participation rule. Discussion can help outside that interval too; question difficulty and prior knowledge may call for a hint or short explanation first (National Academies teaching guidance).

A higher second-vote score alone does not prove understanding: a student may follow a peer’s choice. Ask learners to solve a new, similar item individually and explain why. Smith and colleagues used a follow-up question in an undergraduate genetics course to distinguish understanding from answer copying.

Sources

• Crouch & Mazur, “Peer Instruction: Ten Years of Experience and Results” (2001) — procedure and findings in one university physics context
• Smith et al., “Why Peer Discussion Improves Student Performance…” (2009) — undergraduate genetics study using a similar follow-up item
• National Research Council, “Active Learning in the Classroom” (2015) — guidance on conceptual questions and peer interaction