I'm an upper-division physics student. I have finals coming up next week and I want to take the advice of one of my professors from several years ago: write your own exams for practice and give them to peers. It'll help the test writer to review topics and your peers to have practice questions. However, my finals are all upper-division physics exams. Quantum mechanics, electricity and magnetism, and statistical thermodynamics. Homework and exams are generally pretty difficult, and I'm not sure how I can write an exam that will be on the same level that we can expect on the final.

Does anyone have any suggestions on how to make up questions to create a practice exam for study purposes?

  • Look at exams from previous years? Make sure to cover a bit from every part of the course. Dec 4, 2018 at 20:52
  • Unfortunately we don't have access to those exams. Only one professor consistently teaches the class (QM), I wouldn't be able to get exams from previous years because students who have taken the other two classes are no longer at my university.
    – matryoshka
    Dec 4, 2018 at 20:55
  • 1
    This won't help you now but maybe you could start helping future students. At our university we create memory logs of the exam, right after the exam sit down and write down everything you remember. Maybe talk to friends and split who should remember questions x to y. After the results are published we can take a look at our exams again. Memorize the parts you didn't before. Store these logs in a Facebook group for your studies, shared Dropbox, forums etc. Over the years we accumulated quite and extensive repo of old exam questions for most subjects.
    – idkfa
    Dec 5, 2018 at 8:13

3 Answers 3


For upper-level physics in particular, I would consider using book problems, as it's hard to develop questions that are easy enough to solve in ~15 mins but hard enough not to be trivial or memorization.

  • If you're using a good textbook, then you have likely done quite a few of the problems already for homework / in class, though it's still a good exercise to redo them on the board (make some substitutions where possible) under time pressure and without notes
  • There are also other good books you can likely find in your library: Griffiths or Purcell for E&M; Griffiths or maybe Shankar/Sakurai for Quantum; (not sure about Thermo, I've never found a good thermo book). These will have good problems -- finding answers might be tough, but you can likely figure them out as a group if you choose ones at the right level

This could be in addition to some simple but useful concept questions that you could probably come up with on your own (e.g., what are the implications of the fact that electricity admits a scalar potential while magnetism requires a vector potential?)

  • +1, it's just not realistic to create your own exam questions unless you are very familiar with the material already. Check out past year papers, questions from books, etc. Don't try to set your own.
    – Allure
    Dec 5, 2018 at 4:20

One way to do it is to make a kind of game out of it. Instead of trying to write a complete exam, just come up with one question. Give it to your peers and expect an answer in a few minutes. The first to give a correct answer asks the next question. Each person will need at least a couple of questions to get started, but not a complete exam. Questions asked will, hopefully, help people think of additional questions.

It would be good to do this face to face, but it could be done online if in real time - everyone simultaneously at their computer.

If no one answers in the allotted time, the asker gives the answer and another question is asked, either by the same person or another.

You might be able to use an online forum such as the physics section here to prime for questions. But searching for answers online sort of defeats the purpose of the exercise.


Start by making quizzes rather than exams. Try to come up with True/False and multiple choice questions. Add other types in as well if you want. The point is that you can't make a good question about something if you don't understand it.

I wouldn't bother with sharing it with peers. The important thing is to go over a section and ask, why did the professor bother talking about that section. What was its purpose. If you had to test someone on it, what is the core principle that needs to be tested?

The best thing I have ever received feedback on from students was when rather than giving them a quiz, I made them make and answer their own quiz. It forced them to think about the material. That is all your previous professor was getting at. In my opinion, sharing it with peers is neither here or there. You can do it if you want, the benefit would be to see the mixture of questions you all come up with. But you should also be capable of doing that on your own at this point. Figuring out what is important is the main take away from education.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .