Learn With Jesse
Menu

Group work and discussion

Peer instruction with concept questions

Pose a conceptual question, have every student commit to an answer, let neighbors argue it out, then vote again. It turns a lecture into a series of short, structured conversations.

Moderate evidenceNo AI neededIn personLive onlineAny class sizeA few hours to set up

What it is

In peer instruction the instructor stops lecturing to ask a conceptual question with several possible answers. Students think on their own and commit to an answer (by clicker, phone poll, or flash card). Then they turn to a neighbor, defend their answer, and vote again. The instructor explains the answer after the revote.1

It needs no special room and little class time, which makes it one of the easiest active learning methods to try in a large lecture. It is most useful where students have predictable misconceptions.

Evidence and fit at a glance

Moderate evidence

Consistent results from physics, biology, and other STEM courses, including a two-year college. Strong evidence also supports active learning as a whole. This page did not review a meta-analysis of peer instruction alone, so it is labeled moderate.

How evidence is labeled

  • Set-upA few hours per unit to write or adapt good concept questions. Check whether your discipline has a published bank of concept questions to adapt.
  • Class timeAbout 2 to 5 minutes per question, with 2 to 4 questions per class.
  • DeliveryIn person, Live online
  • Class sizeAny class size. Works best with over 60 students
  • AI roleNo AI needed

When it works, and when it doesn’t

Best when

  • The course has predictable conceptual misconceptions that a good multiple-choice question can expose.
  • The class is large and it is hard to hear from every student, though it also works in small classes with a show of hands or flash cards.
  • Students have enough background to argue about the question. Questions that 35 to 70% of students answer correctly on the first vote are the sweet spot for discussion.1
  • You teach in person, or live online with a poll tool and breakout rooms.

Weak when

  • The content is mostly procedural, with no conceptual question to argue about. Worked examples and practice problems may fit better.
  • Most students answer correctly or almost nobody does on the first vote. Move on to the explanation, or give a hint and try again.1
  • The course is fully asynchronous. Peer instruction depends on live discussion.
  • Students are graded on correctness, which can push them to follow the majority instead of reasoning. Participation credit encouraged more balanced discussion in the reviewed research.1

How to do it

  1. Write a challenging concept question.Aim at a common misconception, with plausible wrong answers. Higher-order questions produced larger gains than recall questions in the reviewed studies.1
  2. Pose it and have students vote alone first.The individual commitment matters. Clickers are not required, and low-tech options such as flash cards performed comparably.1
  3. Look at the first vote.If 35 to 70% are correct, send students to discuss. Below that, give a hint or more explanation. Above that, explain and move on.1
  4. Have neighbors discuss why.Ask them to convince each other with reasons, not just compare answers. Cueing students to explain their reasoning improved the quality of discussion.1
  5. Revote, then explain.Explaining the answer after the revote produced larger learning gains than discussion alone.1

With AI and without

Without AI

AI plays no role. Students answer and argue live, so the structure is naturally resistant to outside tools.

With AI

AI can draft candidate concept questions and plausible wrong answers from your learning outcomes. Questions need expert review, because a flawed distractor or a wrong key wastes the discussion. Keep the live voting and discussion AI-free.

What the research says

What the studies found

A ten-year report from Harvard physics courses for non-majors, both calculus-based and algebra-based, found that students gained in both conceptual understanding and problem-solving skill after peer instruction was adopted.2

Does the discussion itself teach? Students who answered a conceptual question alone, discussed it with neighbors, and revoted usually improved. To test whether that was learning or just copying the best student, researchers gave a similar follow-up question in an undergraduate genetics course that students answered individually. The results suggest peer discussion builds understanding, even when no one in the group initially knows the answer.3

A comparison of a two-year college with a top-tier research university found stronger conceptual learning for peer instruction students and similar problem-solving. Students with less background knowledge in peer instruction reached gains equal to those of better-prepared students in traditional classes, and peer instruction reduced attrition in introductory physics at both institutions.4

Implementation matters

A literature review across physics, biology, chemistry, and computer science found that instructors often modify peer instruction in ways that can reduce its effectiveness. It recommends keeping the individual vote before discussion, asking challenging conceptual questions, choosing when to discuss based on the first vote, explaining after the revote, cueing reasoning, and using low-stakes grading.1

The wider evidence for active learning

A meta-analysis of 225 studies in undergraduate STEM found that active learning raised exam and concept inventory scores by 0.47 standard deviations on average, and students in lecture classes were 1.5 times more likely to fail. The gains held across disciplines and class sizes, with the largest effects in classes of 50 or fewer.5 Lecturing remains the most common format in STEM classrooms, though practices vary.6

Pedagogical justification, ready to paste

Use this in a lesson plan, syllabus, or course-design form. Replace the [bracketed] parts with your own course details. Check the claims against the references before you submit.

One-sentence version

Concept questions with an individual vote, peer discussion, and a revote are used in lecture because students improved in conceptual understanding without losing problem-solving skill in ten years of physics courses,2 and peer discussion builds understanding even when no one in the group starts with the right answer.3

Full rationale

To make large-lecture time active, each class includes two to four concept questions using the peer instruction sequence: students answer individually, discuss with a neighbor, vote again, and the instructor explains the answer. In a ten-year report from introductory physics courses, students gained in both conceptual understanding and problem solving after adoption.2 In a genetics course, students improved on a new individual question after peer discussion, which suggests discussion builds understanding rather than only spreading the right answer.3 A comparison of a two-year college and a research university found stronger conceptual learning and lower attrition in introductory physics, and students with less background reached gains equal to better-prepared students in traditional classes.4 Implementation follows research-based recommendations: an individual vote before discussion, a discussion trigger based on the first vote, an explanation after the revote, and participation-based grading.1 More broadly, active learning raised STEM exam scores by 0.47 standard deviations and lowered failure rates compared with traditional lecture.5

Learning outcome
Students will be able to [explain / predict + concept] and correct common misconceptions about [topic].
Evidence of learning
[Concept-inventory or exam items] targeting the misconceptions covered, plus first-vote and revote counts for each in-class question.
Learning activity
[2 to 4] concept questions per class using vote, discuss, revote, explain, graded for participation only.

The copy buttons paste plain text. The pasted version renumbers the references from 1 and includes them with a source line, so it works on its own. The numbers shown here match the reference list below.

Limits of the evidence

  • Most of the evidence comes from physics and biology. This page does not cite trials in the humanities or business.
  • Writing a good concept question takes expertise and time, and the technique depends on question quality.
  • The 35 to 70% guidance is a recommendation from a literature review, not a law.1
  • The active-learning meta-analysis covers many methods, not peer instruction alone.5 One comparison in a two-year college is promising but is a single study.4
  • Students who dislike group talk may need a clear reason and an easy way to take part.

Questions faculty ask

Do I need clickers?

No. In the reviewed research, low-tech voting such as flash cards produced results comparable to clickers.1

How many students should answer correctly before I start the discussion?

The literature review suggests discussion works best when roughly 35 to 70% answer correctly on the first vote. Below that, give a hint or more explanation. Above that, move on.1

Does peer instruction work at a community college?

One comparison of a two-year college and a research university reported stronger conceptual learning, similar problem-solving, and lower attrition in introductory physics.4 That is one study in one subject, so test it in your own class.

References

  1. Vickrey T, Rosploch K, Rahmanian R, Pilarz M, Stains M. Research-based implementation of peer instruction: a literature review. CBE Life Sci Educ. 2015;14(1):es3. doi:10.1187/cbe.14-11-0198
  2. Crouch CH, Mazur E. Peer Instruction: ten years of experience and results. Am J Phys. 2001;69(9):970-977. doi:10.1119/1.1374249
  3. Smith MK, Wood WB, Adams WK, et al. Why peer discussion improves student performance on in-class concept questions. Science. 2009;323(5910):122-124. doi:10.1126/science.1165919
  4. Lasry N, Mazur E, Watkins J. Peer instruction: from Harvard to the two-year college. Am J Phys. 2008;76(11):1066-1069. doi:10.1119/1.2978182
  5. Freeman S, Eddy SL, McDonough M, et al. Active learning increases student performance in science, engineering, and mathematics. Proc Natl Acad Sci U S A. 2014;111(23):8410-8415. doi:10.1073/pnas.1319030111
  6. Stains M, Harshman J, Barker MK, et al. Anatomy of STEM teaching in North American universities. Science. 2018;359(6383):1468-1470. doi:10.1126/science.aap8892

About this page

Written by Jesse Rogers, Learning Lab Coordinator, Palm Beach State College. Last updated October 15, 2026.

AI disclosure. Model used for research, drafting, editing, and verification: claude-sonnet-5-5 (Claude, by Anthropic). The author is responsible for every claim and citation.

How it was checked. Bibliographic details were checked against DOI records. Numbers and findings were checked against abstracts or open full texts. Where that was not possible, the page says so. See the method page.

License. CC BY 4.0. Share and adapt it with credit. This is an independent project and is not an official resource of Palm Beach State College.

Corrections. If you find an error, message me on LinkedIn and I will check it and fix it.

Cite this page (AMA). Rogers J. Peer instruction with concept questions. Learn With Jesse. Published October 15, 2026. Updated October 15, 2026. Accessed [date]. https://learnwithjesse.com/techniques/peer-instruction/