39

I am a math professor. Several of our midterm questions were posted to Chegg. We found some students who clearly used them and gave them zeroes for the exam. We also had two students where I kept going back and forth. Most of their work was fine but, in each case, there was one answer which was surprisingly close to Chegg but not so close that I thought it couldn't be a coincidence. I kept sitting on the edge and finally decided to let it go.

I've been thinking, after all grades are in, of writing these students and saying "If you were cheating, you came very close to getting caught. You should be scared about how close you came and make sure you don't do this again; you clearly know enough to pass without it. If you weren't, than I am very sorry to have suspected you." Good idea or terrible idea?

20
  • 37
    To me (non US) it sounds very strange that you can deduct points at all for correct answers, just because they were memorized from published solutions online? (They were not using a mobile phone during the exam to copy anything, right?) My feeling is that it is up to the examiner to give suitable questions, and if their questions are identical to ones that have already benn circulated then it is not the student's fault? Or do I misunderstand the question entirely?
    – Jakob
    Dec 10 '20 at 20:02
  • 9
    @Jakob I wouldn't jump to that conclusion given that the sensible interpretation is that the questions were posted and answered during the exam. Which means that copying the solution is clear cheating. No memorizing involved... Dec 11 '20 at 2:36
  • 32
    @Jakob: In the U.S. most of us are currently in a distance-learning environment, and there's no way to prevent students from using desktop, laptop, phone, anything from accessing online answers during a test. E.g., This semester I had a totally unique question, never before posted anywhere, in a discrete math test that was posted to Chegg (incl. my hint for formatting a table in our LMS), received an answer with terms and techniques never covered in class or book, and copy-pasted verbatim to 5 student submissions, all within the 90 minute span of the test. Dec 11 '20 at 3:18
  • 10
    The fact that you want to put an apology in your message to them (not to mention being "very" sorry) should already be a signal to yourself that it's probably not a good idea to send such a message in the first place.
    – NotThatGuy
    Dec 11 '20 at 13:54
  • 5
    @TasosPapastylianou I don't think your assessment is justified. What we're discussing is the following cheating mode: the exam starts, the student posts an exam question online during the exam, someone else (unrelated to the course) answers the exact question and posts their answer online (for a small payment), the original student copies that answer. We are not talking about questions whose answers can simply be found online independent of the exam. Such type of questions would indeed be a bad choice for open book exams (which practically all exams are right now). Dec 11 '20 at 13:57
117

No, you should not single them out. If you have evidence of cheating, then charge it. If you have suspicions then you should swallow and ignore them so as not to prejudice the students in the future.

If you want to announce to the entire class that there were a few other cases that were suspect, not naming names and not singling out any individual, that would be fine and the message would be delivered.

Some people would take such a singling out very badly. Psychologically. Especially if they didn't cheat.

10
  • 89
    @FedericoPoloni if a professor told me that I was suspected of cheating, and I did not, I would be incredibly outraged and pissed. I know for a fact I would be distracted in the course, and constantly worried whatever I did next would be under a microscope (which would be true if I was suspected, but I would be unaware of it). And it would be impossible to look at that professor with hope for any sort of academic relationship. Perhaps in years worth of time, but certainly not in a college span. Others may be more forgiving, but sometimes your reputation is all you have and that stain is deep.
    – J.Hirsch
    Dec 10 '20 at 21:35
  • 15
    Strongly seconding @J.Hirsch's comment... Things/suspicions that are voiced cannot be unsaid. Dec 10 '20 at 21:47
  • 4
    It should also be emphasized that many undergraduate students are under a lot of pressure, especially in their first year, as many of them are experiencing independent life for the first time. Many of them have literally never struggled in school before and have no idea what "studying" is or why you would want to do it. The last thing they need is a thinly-veiled "I think you cheated but I can't prove it" email from the professor (assuming innocence, of course). Obviously, those same students will also have no answer if you demand to see their studying process.
    – Kevin
    Dec 11 '20 at 8:59
  • 15
    I think announcement to the whole class is also profitable towards the people who might have actually cheated without even being suspected. If I was to cheat, and to learn people have been caught cheating, and that others have been suspected and I'm none of them, I probably would feel even more confident. While knowing I might be among the ones being suspected, I might be hesitating next time...
    – Laurent S.
    Dec 11 '20 at 10:57
  • 4
    I can think of one exception to this: the case where it superficially appears that a student has plagiarized a published source, but on closer inspection, it turns out s/he has given proper attribution in some subtle way. I think it's fair play to warn that student "I suggest you make your citations more obvious in future, in case you find yourself falsely suspected of plagiarism." Dec 11 '20 at 14:30
24

If you suspect cheating, one option is to ask the students directly to explain how they got answers that were so similar to the online source. In my experience, this often results in an immediate confession on the part of the students. If they come up with a story that's hard to believe, but you don't have enough evidence to get through the disciplinary process, then you can drop the matter.

6
  • 4
    I completely agree with both Buffy's answer and this answer. Don't accuse students without strong evidence, but depending on how much you care and how sketchy their exam looks, you can require some students to chat with you to "discuss concerns about their exams" before you assign them a final grade. If your university has something like an Academic Integrity Office, you might chat with them first about what to do.
    – Kimball
    Dec 11 '20 at 22:37
  • 2
    Brian, my experience this semester has been the opposite. The email conversation usually goes like this: student asks why they got a zero. I say because they copied the answer to q2 from an online source. They reply that no, that they studied very hard and that all their work is theirs. I send them a screenshot of the online answer they copied. They reply that they were "self-studying", and that it was "the only time", that their life is hard, and that they are in a small village in country X. I send them a screenshot of another answer and report them to the Dean. Dec 12 '20 at 14:05
  • 2
    I've also had students who deny cheating despite evidence showing verbatim copying of large chunks of text. Since the copied text itself is sufficient evidence of plagiarism, I just send the charges on to the dean. Dec 12 '20 at 16:27
  • 2
    @MartinArgerami I think student reactions over email and in person/Zoom can be quite different. If you schedule a conversation with them and ask them to explain how they got their solution and why they wrote such-and-such, some people who copied will confess readily, though it's true that others deny deny deny.
    – Kimball
    Dec 12 '20 at 21:54
  • @Kimball: makes sense. Though in person I've also heard the most outrageous excuses. Dec 12 '20 at 22:16
2

If you ask Chegg, they would supply student data they hold. This includes not only those who asked the questions but also those who viewed the answers. If you have the right clauses in the honor code preventing students from seeking unauthorized help, then you can take students to the disciplinary tribunal with the evidence you obtained from Chegg.

-7

Let's say I'm the student suspected of cheating and look at this from my perspective:

Case 1: I cheated, and receive this email warning me I came close to being caught. This should be a good result. It would act as deterrence - after all, next time I might not be so lucky. Only thing to pay attention to here is to not reveal the evidence against me, since otherwise I might be tempted to cheat again except this time also to cover any tracks.

Case 2: I didn't cheat. In this case the warning is irrelevant to me. Like, the conditional is false, so whatever follows it is irrelevant. After all, the statement "if you cheated then pigs fly" is true! The allegation would only become a nuisance if I had to invest time and effort to fend it off (which I presumably don't in this case), or if it became public (which it doesn't have to, if you send private email).

If you are very worried about case 2 (which is reasonable), you could try looking around to see the effect of false allegations in online forums such as StackExchange. We do for example have rules against multiple accounts. One can never actually be sure that two accounts are actually the same person, which will invariably lead to false accusations and possibly investigations. I don't actually have the statistics on how people react, but perhaps a mod will be able to say. My experience elsewhere is that people get annoyed if they have to mount a defense, but if it's just being told they've been suspected, they might even react with laughter.

Edit: Based on the comments, different people come to different conclusions on this because they interpret things differently. I for example would consider the proposed email closer to being exonerated than being accused, but many others don't. Clearly the dividing line between the two very different reactions is extremely fine. If you do send the letter, you will have to word it very carefully. At that point, it might be better to just sidestep the potential quagmire entirely and not bother.

26
  • 16
    I disagree with case 2. You seem to assume that students think like machines. In reality, I'd assume that most students would intrepret a sentence starting with "If you did cheat" as "I suspect you of cheating" (if the sentence was not given in a complete abstract way or is complete nonsense like your comment). Are you familiar with Schulz von Thun's communication theory?
    – user111388
    Dec 11 '20 at 6:21
  • 18
    @Allure That's not a good analogy because you don't hold any authority over Buffy, and the accusation is pretty silly. Imagine instead that a police man came at your door and told you that have been investigated for some crime at a place that you frequently visit (i.e., you could plausibly be suspected of this crime), but they didn't gather quite enough evidence to charge you. You certainly wouldn't just shrug this off.
    – xLeitix
    Dec 11 '20 at 10:24
  • 10
    I considered adding "just because you say in a comment that it is not an accusation, doesn't make it any less of an accusation" in my last comment. Apparently I should have. You are obviously very detached emotionally from this and believe that it's not a big deal. But it is. For the students it is. A false accusation like that, no matter how you call it, can have devastating effects on the student and might very well cause a lot of otherwise hard-working students to completely stop working for your class, and lose faith in university in general.
    – Stef
    Dec 11 '20 at 11:29
  • 13
    @Allure Clearly you have never been wrongly accused of cheating. It doesn't matter that the accusation is not official, nor does it matter that the teacher adds "if you did not cheat, I am very sorry that I suspected you". Once the words are said, the damage is done.
    – Stef
    Dec 11 '20 at 11:45
  • 8
    No, I have no account other than Buffy (and never have had). You are really mistaken, and insulting. I don't need to game the system here. I just answer a lot of questions.
    – Buffy
    Dec 11 '20 at 14:26
-8

Actually, I wonder why this is a problem at all. Especially math is about understanding and solving a problem. And not about how well you can find external information.

If exam questions are testing understanding and solving, it won't help your students when they are looking up the answers somewhere else. Therefore it doesn't matter if they look up answers. So I think you are fixing the wrong thing here.

Edit: For me it was not clear the questions were posted during the exam. That makes a bit of a difference. When remembering my math exams: besides the answer I had to give an explanation why the answer was correct. Without explanation you got only a fraction of the points. Depending on the teacher, even with a correct answer. When cheating was suspected you had to clarify your explanation. Which you are not able to if you cheated.

2
  • I don't follow the logic here. If a question (which is testing understanding and solving) has been posted in advance online and you memorise and then transcribe the answer in the exam, in what way does it "not help the students"? It helps them to pass the exam without the need to demonstrate any understanding or solving.
    – JBentley
    Dec 11 '20 at 13:56
  • 19
    This question is not about memorizing answers. The exam in question had been posted and solved online during the exam to be shared with the students taking the exam! This is the type of cheating we're talking about. Dec 11 '20 at 14:02

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.