How to Create an Online Quiz for Your Class in 10 Minutes
Building a class quiz usually feels slow for one reason: the work happens in the wrong order. Most people write questions first and only afterwards ask what they were trying to measure. Reverse that, and a usable quiz takes about ten minutes.
1. Decide what you actually want to find out
Before writing a single question, finish this sentence: “After this session, I want to know whether people can ___.” That sentence decides the questions, not the other way round. If the sentence is hard to write, the material itself probably has no clear target — and no quiz built on top of it will measure anything except short-term recall.
For a one-hour session, two or three such sentences are plenty. More than that and the quiz runs long enough that attention is gone before the part that mattered.
2. Pick a realistic number of questions
A rule that rarely misses: one question per two minutes of session for a mid-session check, and ten to fifteen for a closing quiz. A thirty-question quiz sounds thorough, but in a real room people start guessing through the final third — and the data you get back becomes actively misleading.
3. Write questions from real mistakes
The best source of questions is not the course material; it is your own notes about what people habitually get wrong. If every group confuses two particular terms, that is one question. If a procedure step is always skipped, that is another.
Questions born from real mistakes have a useful property: the wrong answers write themselves. You do not have to invent distractors, because the distractors are misunderstandings that genuinely exist.
4. Make the wrong answers plausible
This is the step most often skipped and the one that matters most. Options that are obviously absurd let people answer correctly without understanding anything — they just eliminate the silly ones. Every wrong option should be an answer that someone holding a specific misconception would pick.
A simple test: can you complete the sentence “someone picks B if they think ___”? If yes, B earns its place. If not, replace it.
5. Take your own quiz first
Run through it from the participant screen, not the editor. Three things almost always surface: a question whose answer is given away by the next one, an image that gets cropped on a phone, and a time limit that turns out to be too short to read the question.
6. Read the results per question, not per person
The class average is the least useful number a quiz produces. The useful one is which question was missed most. A single question answered correctly by under 40 % of the room tells you exactly which part of the material needs another pass — and it is often the part you thought you explained most clearly.
In Skorika, the Questions tab shows the correct rate per question while the session is still running, so you can decide to re-explain something before the room empties rather than a week later.
The mistakes that come up most
- Questions that are too long. People read on a phone. If a question needs scrolling, some of them will answer without finishing it.
- Everything at one difficulty. Open with one or two easy questions. People who miss the first one tend to disengage.
- Rewarding speed. If points depend on answering fast, you are measuring reading speed, not understanding. Skorika deliberately gives no speed bonus; time is used only to break ties in the ranking.
- Never retiring a question. A question answered correctly by everyone two sessions running has stopped measuring anything. Replace it.
In short
Write the target first, take questions from real mistakes, make the distractors plausible, test from the participant screen, then read the results per question rather than per person. Those six steps matter far more than which tool you use.