Hacker News Reader

Comment by aqme28 | Hacker News Reader

Is it cheating to study off of past exams? I really don't understand the professor's perspective here.

I think there’s a difference between looking at an old exam and its answers, figuring out why each was right and wrong, and learning from it, and just finding an exam online and using the answer key as your answer in a test because you couldn’t be bothered learning.

The interesting part here is that at least some of the students might have thought it was an old exam from the same course, used it for studying, learned totally nonsensical answers, and then gotten confused in the actual final. Evidently not all of them straight-up copied the answers, or they would have gotten much higher scores. This is apparently still a violation of his institution's academic honesty policy, but there are interesting questions about what happens if he e.g. poisoned one member of a study group who went on to teach bullshit to the other members, giving them a higher chance of picking the wrong answers and putting them over the threshold without ever knowingly cheating.

Of course this is irrelevant, as the p<0.01 test gives an up to 62%[0] chance that someone in his class of 96 students could be falsely accused of cheating, and is unethical in itself. Apparently one student managed to be right on the threshold, which the professor has chosen to interpret as "the student possibly cheating" rather than "me possibly ruining my student's academic life for no reason by choosing a test with lots of false positives and applying it to a class of 96 students".

[0] 1-(0.99^96), and though the real chance is probably a lot lower if the questions are obviously wrong, it's still a stupid threshold to choose

For all of my major exams in the UK, we were encouraged to study by practising on old papers and my score would have been abysmal if I hadn't. So not being allowed to look at past exams feels strange to me too.

Unrelated to the details of this case, but my school had an official, department associated organization/club called the Society of Software Engineers that would run study sessions for CS, SE, and related courses that the students would have to take. They also took in old exams from students to create a knowledge base that you could study against. The rule was that you weren't allowed to leave the room.

It was an interesting grey spot since most professors were all for using old exams to study, but some weren't (I don't think they liked writing new exams every semester, but I wouldn't either). Some would give threats for giving your exams to this university recognized academic club, but they'd black out names upon receiving them so there wouldn't be any proof that it was you (at least easy proof).

It was an interesting case there but most staff were all for it. Making that old data available to future students, but in a more restricted way seemed fine to me, and it absolutely helped me study, but I wonder what others would think of that situation.

It is at the universities whose policies I'm familiar with, yes. At both universities I have attended it was included in the policy alongside all the typical cheating and plagiarism rules.

To some extent yes. By studying the exams, you're learning how to answer the test questions instead of actually understanding the material.

And what is definitely cheating is to search for past exams while you have the current exam in hand and start copying down answers.

Reply on news.ycombinator.com