How to Get Students to Participate During Lectures

Why does encouraging students not work?

A student deciding whether to answer runs a rapid calculation: what do I gain, and what do I risk in front of two hundred people who will still be here next week?

Cooper et al. (2021) surveyed 417 undergraduates and found fear of negative evaluation dominating, and the direction matters. Students are not primarily worried about the lecturer's judgement. They are worried about their peers'.

Severe et al. (2024) put numbers on the texture: about half the women interviewed described not participating out of fear of "looking dumb" against roughly a third of men; nine of ten women raised class size unprompted.

Saying "don't be shy, there are no stupid questions" does not alter any term in that calculation. It is addressed to a motivation problem the student does not have.

So the goal is not more encouragement. It is a cheaper transaction.

What should you say instead of "any questions?"

Participation improves most from changing what you say. These are the specific replacements, and each one lowers the cost of answering rather than raising the encouragement.

Instead of "Any questions?" say: "What would break if we removed the second assumption?"

An open invitation requires a student to decide their confusion is legitimate, articulate it, and accept the room's verdict, all in a few seconds and in public. A specific question with an answer requires none of that.

Instead of "Does everyone understand?" say: "Show me: which of these two is the right method here, A or B?"

The first question can be answered with a nod by a room that does not understand. The second produces a distribution.

How do you get non-volunteers to speak?

Three scripts, and the first is the difference between warm and cold calling.

Instead of calling on someone unprepared, say: "In two minutes I'm going to ask this side of the room what they think. Talk to your neighbour first."

This is warm calling. It gets non-volunteers speaking without the ambush, and the warning is the whole mechanism: remove it and it is cold calling.

Instead of taking the first hand, say nothing, for five seconds.

Most lecturers wait about a second, which selects for students who already knew the answer. Wait five and the composition of the hands changes.

Instead of "Was that clear?" at the end, say: "Write down the muddiest point from today, one line."

A written answer from everyone beats a verbal answer from one person, and it takes the same two minutes.

How do you lower the cost structurally?

Scripts change the transaction. These change the market.

Let them answer without being identified. This is the single highest-leverage change, because it removes the precise barrier the research names. A student who can answer anonymously is not making a social bet.

Let them ask in writing, at any time. Two problems solved at once. The timing: a student's question arrives when they are confused, not when you happen to pause, by which point it has been forgotten or rendered unaskable. The exposure: nobody is speaking in public.

Let the room rank the questions. If students can up-vote each other's questions, non-speakers get a way to participate in setting the agenda, and you get triage for free: the question nine people share arrives first.

Make answering collective. Think-pair-share works because the student reports what the pair concluded. The same sentence carries far less personal risk. Peer instruction is the same mechanism with a re-vote attached.

What should you not do?

Why does the distribution matter so much?

Here is why this matters more than it first appears.

Participation improves learning for the student who participates, through generation, retrieval and timely feedback (Chi & Wylie, 2014; Freeman et al., 2014). If participation is captured by a confident third, then the ordinary lecture is a mechanism for widening the gap between confident and anxious students.

So "get more participation" is the wrong target. The right one is participation from the people not currently supplying it. Every technique above should be judged on that: does it reach the silent two thirds, or does it give the same four students another turn?

What the evidence does not show

• That participation rates are a good outcome measure. They track confidence distribution at least as much as engagement.
• That talking is required. The mechanism is generation and retrieval, not vocalisation. A student writing an answer nobody reads has participated in the sense that matters.
• That response systems raise grades. Hunsu et al. (2016) found a small effect on cognitive outcomes and a near-medium effect on engagement and participation. They reliably change who participates; the grade effect is not established.

Frequently asked questions

How do you get students to participate in a lecture?

By lowering what participation costs them, not by encouraging them more. Non-participation is a cost calculation rather than a motivation problem: Cooper et al. (2021) surveyed 417 undergraduates and found fear of negative evaluation by peers dominating the decision. The changes that work are specific questions instead of open invitations, anonymous answering, written questions at any time, longer wait time and pair-based answers.

Why doesn't encouragement work?

Because it addresses a problem the student does not have. "Don't be shy, there are no stupid questions" is aimed at a belief about the lecturer's standards, while the research identifies the binding fear as the judgement of the two hundred peers who will still be there next week. Encouragement reaches the students who already speak and changes no term in the calculation for anyone else.

What should I say instead of "any questions?"

Ask something with an answer. "What would break if we removed the second assumption?" or "which of these two is the right method here, A or B?" require no admission of confusion, no public self-assessment and no improvised articulation. An open invitation requires all three in a few seconds, which is why it reliably produces silence rather than questions.

Is cold calling a good way to increase participation?

No. It raises the exact cost the research identifies as binding, leaving the student unprepared, named and in public at once, and it is worst for the anxious students you are most trying to reach. Warm calling is the workable version: tell a row two minutes in advance and let them consult a neighbour first. The warning is the entire mechanism.

Should participation count towards the grade?

It tends to backfire. A participation grade converts a social cost into an academic penalty, so a student who cannot speak loses marks for a condition the room created, and it rewards confidence rather than understanding. Widening what counts as participation, so that answering a poll or writing a question counts, addresses the same goal without the penalty.