# It is good to have old exams (with solutions) published?

What are the arguments in publishing or not publishing old exams? If yes, should they contain also a solution? Would you publish them directly after that particular exam or publish your old one's just before a new exam?

The question is somehow related to Is it good to have solutions of homework published?.

At my university publishing old exams is considered a good practice.

• Students know what to expect during real test.
• Students can test their knowledge and check which topics they need to relearn.
• There is an additional problem supply for practice.
• Sometimes professors lose their old tests (e.g. disk failure, etc.), and publishing them might serve as additional backup.
• As pointed out in the comments by Andrew Stacey, it makes sure that everyone has access to tests and so "levels the playing field".

• Creating a new exam requires more work.
• It might be hard to assess the difficulty of a new problem.
• New exams are harder to grade objectively.
• Students know that some areas might be more probable (e.g. theorem XYZ won't be used, because it was given last year) than others.

I hope this helps $\ddot\smile$

• Your first disadvantage is not really valid. Old exams will circulate anyway and so you should come up with new questions regardless. Posting the exams simply makes sure that everyone has access to them and so "levels the playing field". – Andrew Stacey Mar 17 '14 at 11:22
• @AndrewStacey It is true for exams with a few open questions, it's not true for test with 100+ tasks. Also, I've met a teacher who did a two-part exams where the first part was very similar year after year. The thing was, he was asking there about definitions and theorem formulations, and the test was so comprehensive that there was no point in "knowing the questions/answers", you just had to know it all. Anyway, thanks for suggestions. – dtldarek Mar 17 '14 at 12:14
• In fact, the more I think about all of your disadvantages, the more I think that someone who thinks those are real disadvantages has deeper issues than whether old exams are available or not. I would be a little concerned as to what the learning objectives for such a course would be and how they were to be measured. – Andrew Stacey Mar 17 '14 at 12:54
• @AndrewStacey While I would like to agree with you, theory and practice often diverge. It might happen that you have a headache, that one of the questions you have prepared was done in some different class a week ago, there is a group of students that want to take the exam at some other time, etc. It is possible, that in such dire situation reusing an old exam question is a better choice than just inventing a new problem which could be way too easy or way too hard. Of course, it's best to be prepared for any situation, but sometimes it's just not the way it happens. – dtldarek Mar 17 '14 at 13:06
• Publishing old exams helps keep professors honest. It brings to public attention exams that are vague, wrong, too difficult, or too easy. For this reason in some universities it is obligatory to publish exams and their solutions. – Dan Fox Jun 26 '14 at 20:19

Other people have given other good reasons to have old exams be public, but I want to emphasize the one Andrew Stacey points out in comments: old exams often are public, and pretending they're not only confuses yourself and punishes students with less access.

If students get to take their exams home (as they can at most schools, though I know there are exceptions), some organizations will accumulate folders of them to give to future members. (I know of frats of sororities doing this, but other organizations can as well.) Releasing them officially levels the playing field, and maybe more importantly, it makes sure that you remember that students have access to those exams when writing the next one.

• This is actually a major issue at my university. There was a group of students who had access to old exams because they were friends with students who took the class previously and there were a few students who did not. The students who did not have access did significantly worse overall than the students who did. – David G Mar 17 '14 at 18:27

In general, I am in favor of publishing old exams for similar reasons to those given in other answers. However, some caveats are:

• If the course material has changed, questions from old exams may be misleading. Likewise, if the course instructor has changed and the new instructor has a different style of asking questions or likes to emphasize different aspects of the material.

• Students may be overly reliant on previous exam questions when studying and not make sufficient use of other resources to master the material.

• If full answers are supplied, some students might decide to memorize them, in the hope/expectation of encountering similiar questions in the exam.

• Once sufficient old exams accumulate, "learning all solutions by heart" is much more work than learning the subject thoroughly... – vonbrand Jun 21 '14 at 11:51

Exam questions usually are similar to homework questions, but they are not quite the same. Especially for first-year courses, I think it's a very good thing to give an idea how the exam might look like.

If you provide solutions, you will have to decide if you just provide brief answers or full solutions. The former allows the students to check their results, but doesn't help them to learn how to solve this excercise. The latter needs more work, and can give the impression that this solution is the only legitimate way.

In any case, I recommend that there is some way for the students to talk and ask about these old exams and the solutions - be it in the form of a TA session with a tutor, a dedicated part of the lecture or in the form of office hours provided by the lecturer.

• To be able to grade accurately (particularly if several graders are involved) you have to write up a detailed solution. And you need to do it to asess the time required. To publish that isn't much extra work. You can also organize past exams with solutions as a sort of Schaum's for your course... – vonbrand Mar 17 '14 at 16:02
• Solutions which are readable for grading teams don't need to be readable for students - they are written with a different audience in mind. Most of the grading guidelines/solutions I've witnessed as a TA aren't really good solutions for students. – Roland Mar 17 '14 at 16:22
• students must also be able to check if their homework/exams where correctly graded. – vonbrand Mar 17 '14 at 16:39
• @vonbrand: Sure. Still, such a solution to check the grading can differ from one which offers an explanation to the students. Maybe in quality, surely in audience. – Roland Mar 17 '14 at 17:34
• you'll have to provide both anyway – vonbrand Jun 21 '14 at 11:55