Students in our online sessions make memes about the polls.

Not about the content, not about the instructor — about the polls. About the ten seconds of silence before the answer reveals itself. About the moment a “tricky one” appears and everyone scrambles. About the collective groan when the correct answer turns out to be the one nobody picked.

If you want to know whether something is working in a classroom, watch what students do with it when nobody is asking them to. They made memes. That told us something.

Where it started

We run online sessions on Zoom with upwards of 800 students at a time. If you have ever taught to a screen full of face blocks and black rectangles, you know the particular anxiety of it: is anyone still there? Are they following? Has the concept landed, or am I talking to myself?

Zoom polls started as a check-in mechanism. A way of knocking on the door and asking if someone was home. We were not thinking about pedagogy. We were thinking about presence.

But tools, when you take them seriously, have a way of becoming more than what you intended.

The first thing we learned: timing is everything

Early on, we launched polls whenever it felt right. Intuition-driven, loosely timed, sometimes mid-explanation. And something strange happened — the polls that landed mid-thought disrupted more than they helped. Students were still processing the concept when the poll appeared, and the poll pulled them out of that processing rather than deepening it.

So we moved to timed polls. Deliberate placement: after a concept is introduced, before the next one begins. That small change shifted the poll from interruption to punctuation. A well-placed question at the right moment tells the student: pause here, consolidate what you just heard.

Then we started obsessing over the question itself

Once timing was sorted, we noticed the question mattered enormously — not just what it asked, but how we asked it.

Multiple-choice questions with four options require the student to hold and compare. That is cognitive effort, and if the concept is still unfamiliar, the effort goes into reading the options rather than into thinking. We moved toward true/false — deceptively simple, but extraordinarily precise when designed well. A well-crafted true/false question can cut straight to the misconception without the noise of distractor options.

And then there is the difficulty calibration. A question that is too easy produces correct answers and nothing else. A question that is genuinely tricky — one that makes the student doubt themselves for two seconds before committing — produces something richer: that moment of productive uncertainty that good learning lives inside. Students described it as a “high.” They were not wrong.

The memes came from this zone. From questions that surprised them.

The cheating problem — and what it actually reveals

Here is a practical question we spent real time on: how long do you give?

Ten seconds feels tight — and it is, deliberately. Thirty seconds gives enough time for a student to search the answer online. So is thirty seconds a mistake?

We are not sure it is. If a student, in thirty seconds, is motivated enough to look something up — is that cheating, or is that curiosity? The poll has already done something: it has made them want to know. The question is whether the answer they find stays with them or disappears the moment they close the tab.

This pushed us to think about incentive structures. In our programme, participation in polls carries no grade but we did design an incentive structure (saving this for a later piece of Vi prayoga). And yet, 70–80% of students respond on any given day. The design challenge is to make the question worth answering, not to make the stakes high enough to force an answer. Those are very different classrooms.

---

What the responses gave us back

When 800 students answer the same question simultaneously, and sixty percent of them pick the wrong option, it shows you exactly where the understanding broke down. Not after the class, not in an assignment submitted three days later, but inside the moment, while you can still do something about it.

The poll became a live diagnostic. A way of the instructor hearing the room without the room having to raise its hand. In a 800-person Zoom session, nobody raises their hand. But they will click an option. And those clicks, read correctly, can redirect an entire class in real time.

We think of this as the poll serving as an exit ticket — not at the end of the session, but at the end of a concept. Feedback in the moment, for the moment.

The same old question, asked again

A Zoom poll is a simple thing. A few options, a timer, a reveal. Most instructors who use it treat it as a formality — a break in the lecture, a way to check the clock.

We treated it as an experiment. What happens if we change the timing? What if we change the question type? What if we design the question to create uncertainty rather than to confirm understanding? What if we read the responses as data rather than score?

None of these are new questions. How do we drive engagement in a classroom? How do we know if learning is happening? How do we design for curiosity rather than compliance? These questions have been asked for decades, by teachers far more experienced than us, in classrooms far older than Zoom.

At Vicharanashala, we keep asking them anyway. Some of what we try works. Some of it doesn’t. But we keep going — because the question itself is worth it, and because sometimes a tool as ordinary as a poll turns out to be a great pedagogical experiment in disguise.

---

This is part of Vi-Prayoga — where we document the small experiments that shape how we teach.