It was recently brought to the attention of the moderators that a user (who has now been suspended) was posting questions from a take-home test, in violation of the academic policies at his/her institution. I hope we all agree that this is unacceptable behavior, but I would like to explicitly check for a consensus anyway.

  • Do we all agree that this is unacceptable behavior?
  • Should the FAQ be updated in some way to reflect this?
  • 19
    $\begingroup$ +1 It is quite comforting to see moderators soliciting community input on matters of policy. $\endgroup$ Dec 10, 2011 at 3:45

4 Answers 4


Coincidentally, there was another instance of academic dishonesty earlier today, but I was unsure of how to respond to it; I asked the informer how they wanted it to be dealt with and acted accordingly. But this got me reflecting on the issue today...

I am against cheating as much as anyone, but at the moment, my thoughts have coalesced around the notion that our policy should be one of not punishing cheaters via moderator action such as suspension. As a user of the SE network, a cheater has not (necessarily) produced spam or insults, committed sockpuppet voting with their accounts, gone on a downvoting campaign against another user, etc.; in fact they may be a model user of the site in every respect. While it is satisfying to some extent to see them punished in any way, I don't think it's our place to be doing it through the site. It feels somewhat like saying: "someone has been committing burglaries with the use of their car; let's have the traffic police arrange to give them undeserved tickets."

To whatever extent is possible, we should help to make sure they are dealt with in real life (though the moderators and the SE team are unable, as far as I can tell, to assist with any knowledge of their personally identifying information), but that is all we should do. Unless the user has provided identifying information publicly, I think we will have to be content with notifying the course's professor of the existence of a cheater, providing the professor with the cheating user's questions and the posted answers so they can compare with what has been submitted to them, and if requested, the temporary deletion of the user's posts (as was done in this case).

When I think of how I'd like cheaters to be dealt with on the site, I think more of this example. Someone with knowledge of the course catches the cheater, all the cheater's questions are commented on to let other users know not to provide answers, other users are free to express their disapproval with downvotes, and the contact of the professor of the course is organized. Notably, none of it required moderator intervention (though again, when notified we would be happy to help any way we can). Perhaps this would be analogous to having a neighborhood watch in which everyday drivers spread the word to form a car-phalanx in front of the houses known to be on the burglar's list (okay, I have stretched this analogy too far).

As this is meta, I understand completely if readers would like to downvote this if they disagree. I'm not 100% behind it myself; but it is what I am thinking currently. There may well be ramifications of my proposed policy that I haven't thought of, and I would appreciate if people could point them out. I can also imagine that many people may simply disagree with the seriousness with which this proposal takes moderator-inflicted punishment; they might say "the SE network is just a game compared to real life, we should feel free to suspend users in whatever manner we want." I would understand that sentiment as well, and again, I invite people to discuss it here.

Lastly (and perhaps ironically?): I am quite busy with studying for finals and applying to grad schools, so please understand if I am slow to respond.

  • 18
    $\begingroup$ I wouldn't think of suspension in this case as punishment as much as prevention. $\endgroup$ Dec 10, 2011 at 5:34
  • 7
    $\begingroup$ Ah, of course! I see your point. Just as a burglar's car might be impounded to prevent their burglaries, even if they have committed no traffic crimes. $\endgroup$ Dec 10, 2011 at 5:43
  • 5
    $\begingroup$ Or, to use another analogy, you can freeze the assets of somebody suspected to be dealing with shady things. $\endgroup$ Dec 10, 2011 at 5:58
  • $\begingroup$ @J.M. Yes, but can we freeze it forever? $\endgroup$
    – Srivatsan
    Dec 10, 2011 at 17:18
  • 3
    $\begingroup$ @Srivatsan: Well, the user is only suspended for a short time (assuming I've correctly guessed which user it is, anyway); the questions can be undeleted after the suspension period ends. $\endgroup$ Dec 11, 2011 at 0:59
  • 2
    $\begingroup$ This is a remarkably mature analysis, Zev. $\endgroup$ Dec 19, 2011 at 1:18

I would have thought that the second paragraph in the Homework Question FAQ addresses this:

On the other hand, whether your learning institution (middle school, high school, college, etc.) and your teacher or professor allows you to consult other people, or to post the exact question on the internet, is something that is usually addressed by your institution's honor code or rules and regulations, and any specific class policies. You should ask your teacher whether asking a homework question here is appropriate before posting your question.

Of course, it says "homework" rather than "take-home", but here the issue is more a violation of the user's institution academic policies.

I agree that, when moderators are made aware of this actions should be taken (such as suspension of the user).

I'm guessing you are suggesting adding an explicit note saying that the site does not condone willful violation of an institution's or class academic policies, and that if a user is found to be using the site to violate them, then appropriate action will be taken? If so, I'm willing to go with that.

  • 1
    $\begingroup$ There are some subtle issues though. Suppose it is a ver liberal "open book" sort of take-home exam, and suppose the question asks for references on material that would be of help, e.g. background reading. Should we allow that? It may be difficult to figure out where to precisely set the boundary. $\endgroup$ Dec 10, 2011 at 3:56
  • 10
    $\begingroup$ @Bill: I think the key here is that the user was violating the policies of his/her institution. I agree that this is not something that can easily be handled with the FAQ. I think we have essentially the same issue as with homeworks: If the individual's relevant policies say it's okay, it's okay (and users will likely offer help/solutions in much the same way as they do with homeworks; some will argue full solutions are okay, others will argue they are not); if the individual's relevant policies say it isn't valid, then it isn't okay, and moderators should take steps if so informed. $\endgroup$ Dec 10, 2011 at 4:02
  • 3
    $\begingroup$ @Arturo, but we cannot enforce other intitutions' policies! $\endgroup$ Dec 12, 2011 at 15:08
  • 1
    $\begingroup$ @Mariano: No, we cannot bring the student before a Discipline committee. But we can do our part by deleting/locking the questions, at least until the exam is over. $\endgroup$ Dec 12, 2011 at 17:04
  • $\begingroup$ My point is that there are many possible policies at various institutions, so it'd be difficult if not impossible to figure out how to draw a universal boundary that'll work in most cases. Since generally we have no access to particular policies, we have no choice but to take the questioner at their word. Nor do we have any way to verify if the student was informed of such rules, if said rules were presented clearly, etc. Thus I don't think it is correct for us to suspend users for violations of external policies. It's a big can of worms, and our valuable time is much better spent elsewhere. $\endgroup$ Dec 12, 2011 at 22:46

I am sure that nearly everyone agrees that violating the academic integrity policies is ethically unacceptable and if discovered is likely to lead to harsh penalties from the school. The question is not really whether the behavior is acceptable in some ideal sense. The question is what we can do about it, if anything.

I do not think that we should go out of our way to look for these if they are not brought to our attention - we will find false positives everywhere.

The harder question is what to do if a professor brings them to our attention. We very well may not have the means to identify who posted the question, and I don't think we should become a private detective agency.

There is a loss if we delete the question, in that the exam will end soon enough but the question may be of interest to other people later. At the same time, we do not want to give the impression that we are condoning or assisting with that sort of violation.

It seems like a reasonable compromise to me to have a moderator lock the question until the end of the exam, with a note briefly explaining the reason. This still allows others to answer after the exam is over, and it helps to soothe any bad feeling on the part of the professor that we are somehow encouraging cheating. The only loss is that the question does not get answered as quickly as it normally might, but in this situation that would be acceptable to me.

  • $\begingroup$ I don't think this goes far enough. If there are already answers to the question before the question is locked, the OP and any other users taking the same exam can still view them. Better to at least delete the answers (they can be undeleted afterwards). $\endgroup$ Dec 12, 2011 at 7:03
  • 4
    $\begingroup$ "The harder question is what to do if a professor brings them to our attention. We very well may not have the means to identify who posted the question, and I don't think we should become a private detective agency." StackExchange certainly has the means to identify the users (at least by IP address and what not). But per the site's Privacy Policy (see Zev's answer) that information cannot be given to third parties. $\endgroup$ Dec 12, 2011 at 9:52
  • $\begingroup$ @Willie Wong: We can identify the IP address, and maybe an email address, but it is still a long way to go from that to the identify of the person. Particularly if the person tries to obfuscate their identity by, say, using a public hot spot. $\endgroup$ Dec 12, 2011 at 12:24
  • $\begingroup$ @Qiaochu Yuan: I agree with that, I was thinking of questions without answers. $\endgroup$ Dec 12, 2011 at 12:35
  • 2
    $\begingroup$ @Carl: I agree with your conclusion that we shouldn't be a detective agency. But I disagree with your premise that it is because sometimes the "detecting" is hard to do. My point is that the privacy policy precludes us from playing detectives at all, regardless of whether we are capable of doing so technologically. $\endgroup$ Dec 13, 2011 at 9:26

It seems like the questions should be immediately deleted and the user suspended for something like a week, to ensure that the take-home exam is over.

I also think something should be said in the FAQ about not answering questions that look like they're part of a take-home final, and to be especially wary in the months of December, April, and June, when many take-home finals are going on. At the very least, users should be encouraged not to give complete answers to questions that are homework-level, but instead to give hints.

Finally, I certainly don't think that in this case we need to ask what the specific rules of the university are: I can think of no university that would ever condone asking for outside help on an exam.

  • $\begingroup$ I don't think it is healthy for the site to advocate lowering the quality of answers based upon ill-informed guesses about the source of the question. Instead, I think our goal should be to universally strive to give the most helpful answers possible. I would much rather err on the side of (unknowingly) helping a cheater than (mistakenly) not helping an honest student (whose question coincidentally happened to look like a take-home test question). $\endgroup$ Dec 12, 2011 at 22:26
  • 3
    $\begingroup$ @BillDubuque It depends on what is meant by "quality". Some people might consider a carefully constructed answer that gives a student just enough direction to work out the answer for themselves of much higher quality than a blanket full solution. $\endgroup$
    – Matt
    Dec 13, 2011 at 19:55
  • $\begingroup$ @Bill: I was thinking along much the same lines as Matt. In particular when a questioner has shown little attempt to solve the problem themselves, I think that a complete and "perfect" answer could actually have the opposite effect as was intended. These types of answers are better on questions where the questioner shows what they've already tried, and they are the best when answering research-level questions. $\endgroup$ Dec 13, 2011 at 20:41
  • 1
    $\begingroup$ @Matt, Dylan. The above comments are tangential to my main point. Namely, I don't think one should alter the type of answer supplied based on wild guesses about extra mathematical matters such as codes of conduct in some institution (if one even exists - many questions arise based on self-study). I am here - first and foremost - to share mathematical knowledge - not to hoard it. We've already had too many problems arising from some folks attempting to police external matters. That is not the charter of this site. $\endgroup$ Dec 13, 2011 at 21:28
  • $\begingroup$ @Matt, Dylan: In case you haven't noticed Bill is very often just giving well thought out hints instead of full solutions. $\endgroup$ Dec 16, 2011 at 7:02

You must log in to answer this question.

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