Theory
Why Students Don't Ask Questions During Lectures
You ask "any questions?", nobody moves, and you take the silence as understanding. The research is unambiguous that you shouldn't: over half of students in large science courses report never asking or answering a question across an entire semester, and the reason is almost never that they had none.
By Alae Belaich11 min read6 sourcesLast modification 13 August 2026
Does silence mean students understood?
The most expensive assumption in university teaching is that a quiet room is a room that followed you. Two large PLOS ONE studies make it hard to keep believing that.
Cooper and colleagues surveyed 417 undergraduates about asking and answering questions in large-enrollment science classes. More than half reported never doing either across a semester. Not rarely. Never.
Severe and colleagues then went qualitative, interviewing 19 students selected across four quadrants of performance and participation: high performers who never speak, low performers who speak constantly, and both other combinations. They note that in observed life-science courses, fewer than one in three students participate verbally even once.
Put those together and the picture is uncomfortable: the handful of students answering your questions are not a sample of the room. They are a distinct subpopulation, and you are calibrating your pace against them.
Why don't students ask questions?
The interview data is where this stops being abstract. The reasons students give have very little to do with the material.
Fear of being judged by other students. This is the big one. Cooper et al. identify fear of negative evaluation as the dominant factor. Note the direction: students are not primarily worried about you thinking they are slow. They are worried about the two hundred people who will hear the question.
Fear of "looking dumb." In Severe et al.'s interviews, about half of the women explicitly described not participating out of fear of looking stupid, compared with roughly a third of the men.
The size of the room. Nine of the ten women interviewed raised class size unprompted, against four of the nine men. Room size is not a neutral variable; it changes who speaks.
Not knowing whether the question is stupid. A student who is lost often cannot tell whether they have missed something trivial or something everyone missed. Asking risks revealing which.
Speed. By the time a student has decided the question is worth asking, you have moved on two slides.
Why does "any questions?" fail?
It is an open invitation with an unbounded cost. To accept it a student must decide, in a few seconds and in public, that their confusion is legitimate, articulate it, and accept whatever the room concludes about them. Most decline.
It also fails in a way that is invisible to you. Silence produces no signal, so the feedback loop closes with the wrong reading: nobody asked, so it landed. The error surfaces weeks later, in an exam.
Three fixes, in ascending order of effort:
- Ask a specific question instead of a general one. "What's the difference between these two cases?" produces an answer. "Any questions?" produces silence.
- Make answering non-public. Anonymity removes the social cost, which the research identifies as the actual barrier.
- Make everyone answer, not volunteers. A question every student responds to gives you a distribution instead of an anecdote.
What actually gets students asking questions?
Every technique below works by lowering the social cost rather than by asking harder, which follows directly from what the interview data says the barrier is.
Think-pair-share. Pose a question, give 30 seconds of silent thinking, then a minute in pairs, then take answers. It works because the student is no longer speaking as themselves; they are reporting what the pair concluded, which is a far cheaper social act. Deployable with no technology at all.
Anonymous response. Ask the question and let the whole room answer without their name attached. The mechanism is direct: the research names fear of peer judgement as the barrier, and anonymity removes exactly that. This is what response systems are for, and the honest evidence about them is worth knowing (below).
The question queue. Let questions arrive in writing at any time, and let students up-vote each other's. This solves three separate problems: the timing (they can ask when they are confused, not when you pause), the social cost (nobody is speaking in public), and the ranking (the question nine people share reaches you first).
Which changes cost nothing at all?
Two, and they are the ones to try first, because neither needs a tool, a budget or anyone's permission.
Wait time. After asking, wait. Most lecturers wait about a second. Three to five seconds feels agonising and produces markedly more answers. It costs nothing and is the single easiest change on this list. The reason it works is arithmetic rather than psychological: one second is not enough time to parse a question, decide you have an answer, and decide the answer is worth saying out loud.
Warm calling. Not cold calling. Tell students in advance you will be coming to them ("in two minutes I'll ask this table what they think"), so being called on is not an ambush. It gets non-volunteers speaking without the fear that makes cold calling counterproductive for anxious students, because the thing that frightens them is the surprise rather than the speaking.
What the evidence does not show
This is where most articles on this topic overclaim, so plainly:
Response systems do not reliably raise grades. The largest meta-analysis of audience response systems, Hunsu, Adesope and Bayly (2016), covers 111 effect sizes from 53 studies and more than 26,000 participants. It found only a small effect on cognitive learning outcomes. The effect on non-cognitive outcomes (engagement, attention, participation) was near medium.
That is the honest result: clickers and their descendants reliably change who participates and how attentive a room is, and only marginally change what people score. A 2025 systematic review in Humanities and Social Sciences Communications adds a further caution: none of the studies it reviewed used delayed post-tests, so whether any gain survives to the exam is simply unevidenced.
"Under 20% of students raise a hand" is not a real statistic. It circulates widely and is often attributed to a 2024 PLOS ONE paper. That paper does not contain it. The defensible numbers are the ones above: over half never ask (Cooper 2021), fewer than a third participate once (Severe 2024).
The 10-minute attention span is a myth. Frequently invoked to explain silence, it does not survive contact with its own sources. Bradbury (2016) traced the claim and found the primary data does not support a 10–15 minute limit; the largest source of variation was differences between teachers, not between formats.
So: if your room is silent, the research does not license you to blame attention. It points at the social cost of speaking.
What can you change in your next lecture?
| Instead of | Do this | Why |
|---|---|---|
| "Any questions?" | A specific question with a right answer | Produces a signal instead of silence |
| Waiting one second | Waiting five | The commonest and cheapest error |
| Taking hands | Taking anonymous answers from everyone | Removes the barrier the research identifies |
| Cold calling | Warm calling with warning | Non-volunteers speak without the ambush |
| Questions at the end | Questions at any time, in writing | They ask when confused, not when you pause |
Frequently asked questions
Why don't students ask questions in lectures?
Mostly because of fear of negative evaluation by their peers, not by the lecturer. Cooper et al. (2021) surveyed 417 undergraduates in large-enrollment science classes and found more than half reported never asking or answering a question across a semester, with fear of peer judgement the dominant factor. Students are weighing a social cost, not signalling that they understood the material.
How many students actually participate in a large lecture?
Fewer than one in three participate verbally even once in observed life-science courses (Severe et al., 2024), and over half report never asking or answering across an entire semester (Cooper et al., 2021, n = 417). The students who do answer are not a representative sample of the room, which matters if you are calibrating your pace against them.
Does class size affect whether students ask questions?
Students say it does, strongly. In Severe et al.'s interviews, nine of the ten women raised class size unprompted, against four of the nine men. Room size is not a neutral variable: it changes who is willing to speak, and it changes it unevenly across the room rather than uniformly.
Is it true that under 20% of students raise a hand?
No. That figure circulates widely and is often attributed to a 2024 PLOS ONE paper, which does not contain it. The defensible numbers are that over half of students never ask or answer across a semester (Cooper et al., 2021) and fewer than a third participate verbally even once (Severe et al., 2024).
Do anonymous response systems fix the problem?
They fix the participation problem and not the grades one. Hunsu, Adesope and Bayly (2016), covering 111 effect sizes from 53 studies and more than 26,000 participants, found only a small effect on cognitive learning outcomes and a near-medium effect on engagement, attention and participation. Anonymity removes the barrier the interview research identifies, which is a real gain, but it is not a gain in exam scores.