113

Today we (professors + teaching assistants) proctored a midterm exam for a class of about 80 students. There was undoubtedly a "hardest" question on the exam, since nearly the entire classroom of students asked us how to proceed with that question. To be fair, we didn't give any hints, but it was clear that one had to use a definition to be able to proceed.

When we collected the exams, all of the exam booklets had that question unanswered -- except for one student's exam. And this was the only student who asked to go to the bathroom. I gave him permission to go, but I did not ask for his phone, which I now highly regret.

What can we do in this situation, in which I strongly suspect that this student went to the bathroom to look up a definition on his phone, so that he could answer the question correctly?

We still have to look at every exam to be sure that only one student got it right (the bathroom student), but if this were indeed the case, do we have any power to accuse him of cheating, or, have we missed our chance by not asking for him to leave his phone in the room?

I feel a big injustice will have been done to the rest of the honest students, if we let this one slide ...

2
  • Comments are not for extended discussion; this conversation has been moved to chat.
    – eykanal
    Commented Oct 20, 2017 at 13:57
  • 1
    ...nearly the entire classroom of students asked us how to proceed with that question... - what does it mean? Is it ever acceptable for students to ask for hints during the exam?
    – Kostya_I
    Commented Apr 5, 2023 at 9:45

14 Answers 14

315

I apologize in advance for my frankness.

What can we do in this situation, in which I strongly suspect that this student went to the bathroom to look up a definition on his phone, so that he could answer the question correctly?

Nothing.

have we missed our chance by not asking for him to leave his phone in the room?

You missed it, but it doesn't really matter: the student might have had a mate with a phone in the bathroom. Or not. The student might even have known the answer.

I feel a big injustice will have been done to the rest of the honest students, if we let this one slide

In your career you quite probably have already let others slide.

I wrote in this answer that, on the basis of my (now sufficiently long) experience, if a student wants to cheat, they will. Therefore, one has to accept that, from time to time, a student cheats, and one doesn't notice or is not able to prove beyond reasonable doubt that the student cheated (and you don't want to accuse someone of cheating if you're not able to prove it, right?).

So, move on.

NB: As some have remarked in various comments, the fact that almost all students were unable to answer that question can be a major issue. In this answer, I specifically wanted to address the cheating part only, because I think that this other issue is a totally different matter, which can be also culture-dependent, and which is probably worth of a different (interesting) question.

25
  • 236
    I also was a student once, and in one exam I, by pure luck, chose a particularly convenient representation, and solved a question in a super-compact elegant way that no-one, including the professor, had seen before; I had a one-liner response for a question for which several pages of computation were expected. Now, I didn't go to the bathroom; but assume I had, and assume that I had later been falsely accused of cheating, by asking a clever friend while out - how do you think I would have felt? It would have been adding insult (accusation of cheating) to injury (the loss of marks). Cont'd. Commented Oct 19, 2017 at 23:34
  • 158
    Now, I had never been accused of cheating (in fact, I made a point to myself never to cheat), but I was typically known for bloated overcomplicated responses; this compact solution was completely out of character and style for me. So, it could very easily have been that a suspicious tutor would have decided for themselves that I hadn't found elegant solution on my own and canceled the marks. I would have been furious (or demotivated) and definitely would raised a major fuss, if only to clear my name, not even for the marks. No. OP messed up, and without proof, OP cannot accuse. Commented Oct 19, 2017 at 23:40
  • 71
    WRT the last sentence -- Please don't accuse a student unless you have obvious, clear proof, and you've made sure that there's no way it wasn'. I was once accused of cheating because I looked at another student's paper. Pretty clear-cut, right? Except that my vision is so bad I literally cannot see my hand clearly in front of my face (if it's extended more than ~a quarter of the way), so there's no way I could have read text from a student a good eight feet away. It ended up being a massive mess, and caused headaches for everyone. Trust me, it's not something to do lightly.
    – anon
    Commented Oct 20, 2017 at 3:58
  • 37
    This should be the accepted the answer. There is no "calling the student later" or this sort of somehow passive/aggressive things. The exams is the same for everybody and unless there is some clear proof of cheating what is done is done. Next time, be clear about bathroom policy/phone policy. Learning is neverending for both teachers and students, this time your lessons are: -Should I really give questions that can be solved only thru the "luck" of remembering a clearly not-that-central definition? -Should I be more proactive against cheating? Commented Oct 20, 2017 at 8:49
  • 31
    "...if a student wants to cheat, they will." I know someone in IT security who would always say that security systems are mostly intended to deter the casual attacker, but a determined attacker will find a way in eventually. I think the same thing applies to preventing cheating students. I have heard rumours of students developing hand signals or Morse code for communication during exams. There's not a lot you can do to stop the most determined and creative cheaters. Just try to make it so secure that they need to spend more time developing their cheats than it would to study for the exam. Commented Oct 20, 2017 at 19:10
279

Call him in. Ask him how he solved it.

Don't imply that he cheated. Just ask how he solved it with genuine curiosity and interest.

If he asks why you're asking, tell him he was the only student in the entire class who solved it and you'd love to get an insight into his problem solving process.

Either you have a genuine genius on your hands or you have a cheater. Don't do anything yourself to imply the latter possibility.

If he really IS a genius, you will be glad you had this interview with him and can now give him the scholastic accolades he deserves! Rather than going on falsely believing he is a cheater.

If, as you suspect, he was cheating, he may "fess up" to it. Or he may not. But if he can't explain the answer or how he arrived at it in any way, shape, or form, you will know he was cheating. And he will know that you know.

And that may be enough to change his behavior for the better in the future.


If it turns out he was cheating (either by his admission or by his total failure to be able to explain anything about the answer), the lightest thing you could do would be to not give him credit for that question. So then no one gets credit for it, since no one else answered it. Or, as others have said, you might just let it slide as far as grading is concerned.

But having the above interview gives you the chance to (a) discover a possible genius or (b) put a little bit of discipline on the student so he's aware he didn't "get away clean" and will be less likely to cheat again.

The very least it will do, if he cheated, is to make him sweat.

1
  • Comments are not for extended discussion; this conversation has been moved to chat.
    – StrongBad
    Commented Oct 23, 2017 at 14:37
86

Not exactly an answer, but perhaps some perspective: a long time ago, I went to one of the U.S. service academies (Army, Navy, Air Force, Coast Guard, nevermind which) and, like all incoming people, was subjected to an extensive battery of placement tests, some generating college credit, and so on. Far more systematic than most U.S. colleges and universities even today.

Having been an avid student of mathematics for some years prior, I had indeed read a lot, and in particular had known how to do calculus (which was the basic entry-level topic in mathematics) for some years prior. While doing generally well on the other exams, apparently I only mis-answered a single question of 100's on the calculus exam. Since even competent people rarely do that well, and since I did not have high school calculus on my record, this was suspicious. (Let me remark that, yes, I did in fact know how to do calculus pretty well, and, yes, I also knew how to "game" multiple-choice questions. And I guess I had a good day, too.)

A point is that at the U.S. service academies, any lying or cheating or anything-at-all is a dismissable offense. So, given the suspicious nature of the situation, I was called before an "officers board" to account for the situation (since there was certainly no overt evidence of cheating).

Being a naive, scared kid standing at attention in front of officers, to the question "how do you account for this?" my initial response was "Sir! I read a lot of books! Sir!" (That was the required style of address...)

(This got a laugh, which did not calm me, by any means, because at the time I didn't know how to interpret it.)

The wrap-up was that I was not punished or dismissed from the place...

So: in your situation, you'd definitely need to interview the student before thinking in terms of accusations. Some people are not typical, even though, yes, statistically they are.

Also, the whole "deny bathroom break" thing is silly, and insisting on "accompanied breaks" is silly/rude, and so on. The complications to "testing" are not solvable by outlawing bathroom breaks.

Nor by outlawing phones, because dedicated cheaters can get much smaller devices...

"Catching and punishing cheaters" is obviously not the primary goal of education of any sort, so we don't want to let that goal corrupt the rest of it.

5
  • 3
    Unfortunately in the military they hold a lot more cards than a teacher (particularly public school). While it's not actually constitutionally so (at least in the US), public education is seen as being basically a right. And any misstep is under the guise of a government entity. Such that lawsuits are even a more common event. At the least, most schools are now quite rapid to dismiss teachers at even the hint of an issue that may cause media attention. One student gets "interviewed" and is "fragile", you've got a crying kid on tv, and a mess. It's a world of scary caution anymore Commented Oct 20, 2017 at 13:58
  • 29
    And in this world, a cartoon like i.pinimg.com/originals/5b/91/09/…... hits far too close to home :-/ Commented Oct 20, 2017 at 13:59
  • 1
    Oh, and while in no way am I suggesting that the very vast majority of sexual assault allegations are anything but true... it's nonetheless real that students who are willing to cheat may be willing to go to other means to get their way. Places like military academies will generally quite naturally filter down to be those motivated by a purpose, and ethics. But at public schools (even many CCs/universities) some are there because they're told to be/fairly aimless, and may be uninterested with morality not seeing any benefit to them... and so may seek revenge or means to restore control :-/ Commented Oct 20, 2017 at 14:15
  • 9
    I once was that student as well. Failing nearly every easy question, but succeeding the hardest one(s). The first time it was suspicious. The second time, it became even more suspicious. Since then, I had a personal supervisor during my exams. Until the third and fourth times came. The supervisor didn't find any cheating. I was only bored and wanted challenges: I overlooked easy questions and focused on hard ones. I became known as the one who solves only the hardest questions and teachers left me alone. That doesn't make me a cheater: the system was inadapted for me (or vice-versa). Commented Oct 21, 2017 at 14:30
  • 1
    @OlivierGrégoire, off topic, but you would love the book Concrete Mathematics if you haven't discovered it already. It has virtually nothing in it but interesting, challenging questions.
    – Wildcard
    Commented Oct 26, 2017 at 20:06
52

If you can't prove that a student cheated, you can't prosecute or penalize them. Bear in mind cheating is a very serious charge, and can ruin a student's academic career.

Don't throw this charge around lightly.

If you're the teacher, and if your test isn't well-written enough to be able to discern if someone knows the material (v.s. googling the topic in the bathroom for 5m), your test-making skills could stand some improvement.

In other words, rethink your student evaluation process if it can be defeated by a cursory google search, or a 'cheat-sheet'.

1
  • I am not sure if I agree with the thesis that professors have to pay a price (rewriting exams to be ungoogleable) simply because students cannot be trusted to abide by the honor code that they signed. It's an interesting question. I prefer an explicit policy that gives the professor the right to ask the student to explain a solution, and gives the professor the right to change the grade if the student cannot demonstrate an understanding of the work they handed in. Commented Aug 16 at 13:42
20
  • First, a bit of (my) perspective: Don't overestimate the significance of grading people, in general. As a researcher I have not found that much use for the exam grades of undergrad/grad students I encounter. It's basically something the capitalist economy needs, or thinks it needs, for employee selection. This, as opposed to giving students feedback on their answers - what they got right or wrong and where their mistakes seems to stem from; doing that is super-important. So the fact that the bathroom-student's grade may be wrong, in itself, would not seem to me like the sky is falling.

  • On the other hand, fairness is a big deal in my book. Thus the prospect of someone having gotten phone help on a hard question and "sticking it" to his/her fellow students would bother me. However

  • Either the question you gave was solvable - in which case you have no basis for suspicion - or it was objectively practically-unsolvable, in which case I don't believe you even have a moral leg to stand on for accusing the student of misconduct. To quote the line from the biblical Samson: "If you had not plowed with my heifer, you would not have solved my riddle" - if it were a question which half the class answers, you never would have been suspicious of the "bathroom student".

  • You should adopt the proposal of @SolarMike and just cancel the question, so that the grade is based on all of the other questions and this one doesn't count. If that student complains, you can figure out whether he actually understands the material well enough to have solved the question him/herself... also, +1 @SolarMike's answer.

  • Next time, Have the least experienced TA in the course, who had not seen the exam in advance, sit down and solve it - before administering the exam to the students. That will save you a lot of grief if you can manage it.

  • The greater injustice in this case is you guys either writing an inappropriate exam question or failing to educate your class to answer a question on an appropriate exam. Try focusin on bettering yourselves rather than on punishing the misdeeds of others.

8
  • "- or it was objectively practically-unsolvable, in which case I don't believe you even have a moral leg to stand on for accusing the student of misconduct. " I do not see your logic. Commented Oct 21, 2017 at 21:41
  • 5
    @Acccumulation: It's like an abused spouse shooting their abusing wife/husband. Sure, shooting people is wrong, but the abuser doesn't have a moral leg to stand on and complain about having been the victim of this kind of violent act. If an unsolvable exam drives someone to cheat it is not for the person who administered it to complain. Ok, that's not a perfectl analogy, but you catch my drift. Or - read the Samson Riddle story.
    – einpoklum
    Commented Oct 21, 2017 at 22:35
  • 2
    Giving hard questions isn't unethical, and it doesn't force people to cheat; anyone who cheats in that situation is making that decision on their own. Your analogy relies on the premise that the professor was unethical, which is the whole issue at hand. Analogies are for explaining claims, not for simply reiterating them. I'm not clear on what relevance you think the Samson story has. Is the professor Samson? Commented Oct 22, 2017 at 0:31
  • @Acccumulation : Yes. More broadly, Samson represents whomever is innocent and whom the wrongdoing was against. Which in this case I equate to the professor.
    – TOOGAM
    Commented Oct 22, 2017 at 0:54
  • 5
    @Acccumulation: If nobody found an answer, the question wasn't "hard". Also, no, the student is Samson - which was indeed cheating with his riddle; but the Philistines couldn't have known that without their own inappropriate conduct.
    – einpoklum
    Commented Oct 22, 2017 at 1:07
15

Just remove that question from the paper - on the grounds that so few managed it or the material had not covered that sufficiently...

2
  • Comments are not for extended discussion; this conversation has been moved to chat.
    – StrongBad
    Commented Oct 23, 2017 at 14:39
  • I often feel like a nonlinear scale might be a good idea, or be able to earn bonus points (i.e. answers add up to >100% if you get them all correct). No need to dismiss it just because only one person got it correct (especially if >0 got it correct). If someone genuinely studied very well, had some bad luck with the other questions, and got the bonus one(s) right... that should be good. And anyone scoring >100 is just capped to 100%.
    – Luc
    Commented Oct 23, 2017 at 22:20
7

Disclaimer: I have never to date taught a class and never compiled, overseen or marked an exam.

As much as I believe cheating or attempting to cheat is a bad attitude and as much as I want cheaters to be punished accordingly, I also believe that proving cheating must come down to hard evidence. Hard evidence meaning enough evidence that you would win a criminal case in court if it were taken there. One key principle is in dubio pro reo and this applies here: you do not have proof, you only have a hunch. Unless you manage to find actual, factual evidence do not accuse this student of cheating and do not adjust only their mark in a way that suggests cheating.


I for my part know that I sometimes remember the weirdest details while missing out on often-repeated details. So I might well have performed poorly or averagely in that exam because of not knowing general stuff — but by chance I may remember exactly that definition almost word for word (or symbol for symbol if this is a mathematics exam) and be able to answer that question.

9
  • 2
    Cheating isn't as serious as murder, so it is acceptable for the burden of proof for the former to be lower than for the latter. Commented Oct 21, 2017 at 21:40
  • 9
    I hereby accuse you, @Acccumulation, to have insulted me. Now, insulting someone may, depending on the details, be less severe than exam fraud. Therefore, the burden on proof is on you. In case you cannot prove your innocence, German courts grant me some amount of money up to 2,000 EUR. Yay :) Oh, while we're at it, I accuse you of illegal disposal of oil. And because it isn't murder, the burden of proof is on you, too.
    – phresnel
    Commented Oct 23, 2017 at 9:55
  • 1
    @phresnel I think the "insult" example is better than the oil spill. The damage is "virtual", and therefore the similarity is much clearer. Apart from that, I agree with your point. Commented Oct 23, 2017 at 10:57
  • 7
    @Acccumulation I'm going to say that Phresnels arguments were very likely made in good faith. They're a bit cheeky, but the ridiculousness of them was presumably intentional to show that (s)he was not serious in the accusation, but instead trying to point out the problems inherent in a sliding scale for the necessary evidence for the burden of proof.
    – Beska
    Commented Oct 23, 2017 at 13:48
  • 1
    @Acccumulation In the US at least, all criminal cases must be proven beyond any reasonable doubt, just like murder. Then again this might show up in civil court if handled unfairly, only with the university as defendant!
    – user45501
    Commented Oct 24, 2017 at 8:05
5

I pretty much agree with Massimo Ortolano's answer, but there might be one thing you can do. Run a search on the web with some sentences of the suspicious answer. If the student copied verbatim some text available on the net (something more extensive than a single generic sentence), then while you won't be able to prove he or she cheated, you will be able to prove plagiarism. That may, or may not be useful.

I had a similar problem once in a course preparing to a national competition. At one test that was taken from a former competition sheet, three papers where copied extensively from a correction of that test available on line. The students argued they learned the solutions by heart to several of these competition sheets, which might be true (I don't know if it would makes me more sad if that where true, or if they had cheated with phones and lied to our faces). I tried to make them understand that plagiarism was a serious offense even without cheating involved, but we did not proceeded with the disciplinary board. That test was not used for grading the course at all, and I explained why to all students (without giving names) -- this was possible because we train student with numerous such tests.

2
  • There are people with exceptional memories. And knowing solutions to known problems for such people can be like knowing vocabulary of a language: an anchor. Neither punishable, nor a legitimate reason to deplore anything, but a perfectly legitimate way of acquiring knowledge. It does not even mean that these are not original, as they may save their creativity for other problems. Commented Dec 31, 2019 at 9:40
  • @CaptainEmacs: In my 2nd paragraph, the point was not to punish memory, but to punish plagiarism - if one memorizes a solution and uses it word-by-word without using quote marks and mentioning the source, then one commits plagiarism: using other's work under their own name. If they memorize and understand the answer, they are free to rephrase it in their own words. Commented Jan 3, 2020 at 14:56
2

If the question was really too difficult for 79/80 students to attempt, you should consider not punishing them for missing the answer.

Note that this is not because 1 student did answer it, but despite that one student did answer it.


Practically the implementation may be the same (e.g. everyone gets full/partial score for the question), but because this is about ethics it is important that you only do this if it is not to punish the 1, but because you want to be fair and the question was too difficult.

2

What can we do in this situation, in which I strongly suspect that this student went to the bathroom to look up a definition on his phone, so that he could answer the question correctly?

You can examine your suspicions:

  1. Is the student’s conduct suspicious? What conduct? At worst, the student went to the bathroom, with permission, without surrendering a posession that they weren’t asked to surrender. Not suspicious. At best, they didn’t have a phone to begin with.
  2. Is the student’s answer suspicious? No, based on this question.

Now you say that it’s suspicious because the student was the only one to answer it. But you must have expected that someone would be able to answer it, or you wouldn’t have included it, right? And since you are a competent teacher, it had might as well be one of your students, right?

So the answer is: Unless you can come up with some better evidence, do nothing.

(Or, if your university has an integrity board, let them handle it. In that case, you shouldn’t even be here asking about it.)

0

Once, a long time ago, in an elementary school in which I went, we had an exam (I do not know the English word for an exam in elementary school, it could be a word "test", but I am not sure) in physics, and, as far as I can remember, I was the only pupil that solved the hardest question on the hardest version of a test (I think there was some classification of tests, depending on maximum grade that can be obtained, one class of tests had some maximum grade, another class another one and so on, maybe there were only two classes of tests).

Also, it seems to me now that only I solved that question not as only in my class but as only in the school, because there were some pupils that approached me like : "How were you able to do that? No one did it."

Well, I answered the hardest question on the hardest of exams (tests) because I remembered some scene from a movie I watched, where an actor while being in wilderness and trying to do everything he can so as to save his life, did exactly what was the answer to the question of mine. There was almost no way to answer that question by only going through what was given to us from our teacher before.

I did not mention some also important facts here, but it could be the case that bathroom student solved the question without any cheating, if I were you, I would talk to him and explain to him the whole thing, I would tell him that if he cheated in any way that he should admit that so that he and you feel better, and if he cheated and is able to admit it, I would not tell to other students that he cheated, and would not view him because of that as more bad than some others, if he really cheated and is able to admit that then he only needs a good talk and a clarification of what exactly did he do. I wouldn´t also, if he cheated, accuse him or un-reward him, or take some of his points from the exam, I would just explain to him that what he did is not right, and that´s all.

1
  • -1 This answer looked good, until this: “I wouldn´t also, if he cheated, accuse him or un-reward him, or take some of his points from the exam, I would just explain to him that what he did is not right, and that´s all.” No, that’s not how it works. Not in any university I have heard of. Commented Nov 18, 2020 at 14:26
0

If the method offered by Wildcard doesn't help you determine whether the student cheated or not, your best bet is to either let it go, or find a way to adjust how you weigh the entire test in the final grade.

You may consider dropping each student's lowest test score (for example). You can even drop each lowest-scored question on this particular test (assuming all questions have the same point values).

However, (since you are troubled enough to ask for help here) I assume you understand that you cannot accuse a student whom you suspect if you don't have the proof to justify the accusation.

0

I am puzzled by the fact that this "hardest question" could hinge on just knowing "a definition." A pragmatic approach would be to exclude this question from the exam, on the grounds that with 79 out of 80 miss rate it was clearly inappropriate given the reasons we organise these exams, and rescale the remaining questions.

-3

If you think that you failed to teach the class sufficiently enough for any your students to be able to answer the question on an exam you gave them, then the things that are unfair are these, and a few others: 1. You put questions in the exam that you did not teach them adequately 2. You assume that your student cheated

The exam and its results have no value in the real world. The knowledge gained from a well-taught class has value. The skills learned while studying subjects has value.

Do a much better job teaching the subjects. Do a much better job building exams that test the students on only the material that you sufficiently taught them. Grade your teaching skills on the results of the student's ability to answer the questions.

If you had the only student who realized that it was completely unfair of you to have put a question that you as the teacher thought not one student should have been able to answer, and that student decided to make it fair, by looking up the answer themselves, then that student deserves extra credit.

You are not the only teacher who has their perspective on what matters wrong.

Praise the student for being the only one to get it right, and ask them about how they got the answer.

5
  • 8
    Deciding to cheat is not an appropriate answer to unfair exam questions. Two wrongs don't make a right.
    – nengel
    Commented Oct 20, 2017 at 2:45
  • 2
    I understand the sentiment you are expressing regarding unreasonable exam questions, but it's quite a leap to suggest that would somehow validate cheating (if that is even what happened here).
    – Folau
    Commented Oct 20, 2017 at 10:16
  • 2
    Condemnation of the teacher solves nothing. However, if I were the teacher, I might (disingenuously) say something like, “Congratulations to the one student who got ___ correct. Since everyone else had trouble with it, it’s possible I didn’t teach it. So the exams will be scored as if they did not include that question.” So the student now knows that IF he cheated, it was to no avail. And if you catch his eye while you are saying it, a very subtle change of expression will make him think you suspect cheating.
    – WGroleau
    Commented Oct 20, 2017 at 10:57
  • 10
    WGoleau, that's unprofessional behavior. Students are no less intelligent than teachers, just less knowledgeable. Let's not use our podium as a chance to play "clever" word games with hidden accusations. Commented Oct 21, 2017 at 23:40
  • @JosephSimmons I've had multiple teachers say exactly that. It's not a hidden accusation, it's just the standard way of handling an improperly written quiz. If only one person got the answer, they would congratulate them (give them candy sometimes) then say they were making it extra credit, or non-graded. I've probably seen it happen about 20 to 30 times in High School alone. They didn't do that sly expression at the end, however.
    – user78960
    Commented Oct 22, 2017 at 17:47

You must log in to answer this question.

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