Interact
Grading a quiz in person always raises the same uncomfortable question: how can you be sure the score reflects what each participant truly knows, rather than what their neighbor answered two seconds earlier? You are not going to confiscate phones or patrol the aisles. The right answer is not surveillance, it is design. A well-built quiz makes cheating slow, tedious and pointless, without you ever looking up. This guide walks through the five concrete settings that make the difference, and how to switch them on in a few minutes on your Mac.
Why cheating is not a surveillance problem
Most in-room cheating is cheating by convenience: a glance at the neighboring screen, waiting for the room to react before committing, copying an answer left in view. It works because everyone sees the same question, at the same moment, in the same order, with the same options in the same place. Break that synchronization and copying becomes work: your neighbor is not on the same question, their choices are not laid out the same way, and they no longer have time to help.
In other words, a reliable graded quiz is not defended by watching people, it is defended by its settings. It is also more respectful: nobody feels policed, and you stay in your role as a facilitator. If you first want to understand what makes a quiz truly measure a learning outcome, the article a quiz that really assesses lays the groundwork; this one is the anti-cheating how-to companion.
You do not supervise a quiz, you design it so there is nothing left to supervise.
Setting 1 — Assessment mode: everyone at their own pace
The first lever is to decouple the participants. In assessment mode, each person moves through their own series at their own pace: there is no shared "question 3" on the big screen that the whole room attacks at once. When everyone is at a different point in the quiz, copying your neighbor no longer helps, because they are not where you are.
This mode also shifts the teaching dynamic: you leave the collective speed game and enter a genuine individual test. Nobody is punished for thinking three seconds longer, nobody clicks in a panic because a shared countdown is running. The score gains accuracy because it measures understanding, not reflexes.
Setting 2 — Randomized question order
The second lever, and probably the most effective: randomize the question order for each participant. Two people side by side never see the same question at the same instant. The classic whispered "what did you put for B?" falls apart, since one person’s B is another’s F.
This random order carries a quiet side effect: it neutralizes passive help. Even with no bad intent, a participant who overhears a snippet of conversation cannot attach it to their own question. Each paper becomes truly individual, without you adding a single visible constraint.
Setting 3 — Shuffled answer choices
You shuffle the question order, but also the order of the options inside each question. The correct answer is not always "the second one" or "the middle one". As a result, memorizing a grid like "1-C, 2-A, 3-D" between two sessions becomes useless, and copying the position of a neighbor’s click gives no information.
It is a detail that costs zero effort to enable and closes an entire door. Combined with randomized question order, you get as many versions of the quiz as there are participants, without writing a single variant by hand.
- Assessment mode: everyone works through their own series, at their own pace.
- Randomized question order: two neighbors are never on the same question.
- Shuffled choices: the position of a correct answer no longer means anything.
- Per-question timer: no time to reach for a phone or look elsewhere.
- Automatic score out of 20: the grade lands on its own, no manual marking.
Setting 4 — The per-question timer
A overall time limit for the quiz does two things at once. First, it stops the participant from going to fetch the answer elsewhere: firing up a search engine or digging through notes costs more seconds than remain. Second, it standardizes effort: everyone gets the same window to answer, which makes the score comparable from one person to the next.
Calibrate that delay sensibly: enough to read and think, too short to go cheat. A simple recall question fits in fifteen or twenty seconds; a reasoning question deserves more. The timer is not there to stress people, it is there to set a frame. If time management interests you beyond the quiz, everything is set from the same interface, including the session’s projected timer.
Setting 5 — The automatic score out of 20
Once the quiz is over, each participant has a score out of 20 computed automatically from their answers. You do not go back over it with a grid and a red pen: the scale applies on its own, immediately, with no transcription error. This is where anti-cheating meets time saved, because a reliable score that lands by itself is also marking you will not redo in the evening.
These scores stay useful: you see who fell behind on which concept, and you can move on to targeted remediation instead of a fuzzy average. That is exactly the spirit of formative assessment — the score does not close the sequence, it tells you what to rework.
Where your data lives, and the rest of the setup
This whole setup runs on your Mac. Participants join the quiz by scanning a QR code, with no app to install and no account to create, and the quiz is one activity among the 14 available in SnapInteract — so you can chain it with a poll, an idea wall or a document drop-off in the same session. On the privacy side, your data stays on your Mac: we keep nothing on our servers.
To go further, take a look at the full catalog of 14 activities and the pricing page: the free plan is very generous, with a monthly cap on interactive activities and the rest unlimited, and the Pro plan at 9.99 EUR/month (or 59.99 EUR/year) unlocks everything. Enough to build a first graded quiz, try it under real conditions, and see the difference on the papers.
Frequently asked questions
Does assessment mode really stop people from copying their neighbor?
Largely, yes. Everyone works through their own series, in a different question order and with options arranged differently. Your neighbor is almost never on the same question at the same moment, so glancing at their screen gives you nothing usable.
Do participants need to install anything to take the quiz?
No. They scan a QR code shown on screen and the quiz opens in their browser, with no app and no account. Nothing to download, nothing to create: they are in the quiz within seconds.
Is the score out of 20 calculated automatically?
Yes. As soon as a participant finishes, their score out of 20 is established automatically from their answers. There is no manual marking to do, and you get a per-concept view to target remediation.
Is this graded quiz available in the free plan?
The quiz is one of the 14 interactive activities. The free plan allows a certain number per month through a monthly cap, with the rest of the app staying unlimited; the Pro plan at 9.99 EUR/month unlocks everything if you grade often.
