OrangePhilosophy

There is another option.

What you do is:

(1) find two questions (two on each of their tests) that most closely match up with their counterpart on the other student's test.

(2) determine what the answers are worth.

(3) On student As paper you mark question 1 up five points and question 2 down five points.

(4) On student Bs paper you mark question 1 down five points and question 2 up five points.

(5) wait and see how enterprising the students really are.

I don't think that this solution is wrong (morally) at all.

You are there to teach. Being examplary for you lack of cynicism or for liberally crediting all efforts is not strong teaching.

Teaching that cheating is noticeable and punishable it a very imporant lesson.

Teaching that amibiguous situations default to the credit of the student [benefit of the doubt] is an ambiguous teaching and one that is bound to land you many more such grading quandries.

The solution Mark B poses might clear up several questions, one in your mind and some in the minds of the possibly-collaborating students....you just have to allow for mistaken assumptions on the part of all parties including yourself. Its not a miltary operation so do-overs or other wiggle room may serve.

Jeremy,
hey (I'm a grad student at UCONN).

I'm not sure if what you suggest, or if what Mark suggests, is morally right or wrong. It is a difficult one to determine. I would suggest it is much better to err on the side of what would clearly be the RIGHT moral action in this case.

Namely, fail both students entirely. Give them each a zero. Offer no explanation initially. Then, when they both come to you to complain, make it their job to prove to you that they didn't cheat. The burden of proof is on them. If they prove their case effectively, give them the score they earned assuming there was no cheating. If they do not prove their case satisfactorily -- fail them. Academic integrity is an area where we cannot bend in any way whatsoever. It's funny, because in most ethical situations I err on the side of mercy. But in the area of Academic integrity -- especially in this day and age -- the "lesson" taught to these students would be more important and better than most others they'll ever learn. Espcially when you look at the statistics that show cheating has been dramatically on the rise in the past 10 years -- all the more reason for you to make a bold statement. What do these things we call "degrees" even mean if a great number of students now cheat their way through to get them?

Again, if they can prove to you they did not cheat -- great. But if they can't -- zero is the grade they get. Just my two cents.

Cheers.

My wife told me that a professor she TA's for has a very Machiavellian approach to these kinds of problems. Today she went in front of her class clearly a little steamed. Then she told them that they had discovered some cases of cheating. She said that cheaters who contacted her by Friday would simply fail the class. However, if they failed to come to her she would press their case with the judicial review board, and advocate that the persons be expelled from the university.

Now truth be told, they knew of one case of cheating, but by 5pm six people had emailed to confess and a couple had come by the office.

The suggested course of action strikes me as cowardly. The evidence you cite sounds drastically insufficient to warrant a charge of cheating. (For all that's been said, both having taken good lecture notes and then working from them would explain the similarities cited.) Either explicitly confront them with your suspicion and let the ensuing conversation inform your decision on whether to pursue the matter further, or grade them according to the merits.

If they were smart enough to disguise their ruse the way you suggest, they likely aren't going to be dumb enough to come to you and complain about a few points difference. And in any case, you

The suggestion that they "prove they didn't cheat" is preposterous, given the meager evidence cited.

I agree with the last comment by John, which is why I didn't do anything about it on that exam. I wasn't sure if my proposal was immoral, but I'm absolutely Jay's is immoral. I can't fail people if there's insufficient evidence to prove cheating.

I do think there's a stronger reason for doing something at this point, but that's based on a second instance with even more similarity. I think I'm considering doing something along the lines of Matthew's case (but with lower penalties). Since it's still possible they both took good notes or shared their class notes with each other (which I told them explicitly they could do), I think I might announce to the class that I found some very similar exams on the second and third exams, and I let it go on the second exam for the reasons stated in the previous paragraph. If the people come forward, I'll let them do the exam again. Otherwise, I'll fail them for the third exam.

If, as a professor, one wishes to have no collaboration on examinations, the examination should be given in a controlled site where collaboration is not possible. One of my graduate professors (Systems Management) stated that, in his opinion, collaboration was not only expected but essential given that we would spend our future lives in collaborative situations.
Given my statement above, I believe that there is insufficient evidence to prove cheating has occurred and that cheating may not have occurred even if there was collaboration. The answers should be graded upon their content and logic. As in business, philosophy is rarely a single person endeavor so collaboration coming up with well reasoned answers should be the goal.
Finally, I am disturbed by the answers that state the students are required to prove innocence. One of the foundation blocks of the philosopher/constitutionalists who founded this nation is that one is innocent until proven guilty. Being guilty until one proves innocence is a tenet of dictatorships and other similar governing agencies.

Ed

Jeremy,

Last year I had an incident involving cheating and I made the sort of offer that Mullins mentions above. Looking back, I think this was a serious mistake since it creates an incentive to (quasi) confess to cheating that didn't occur. I'm not well acquainted with the literature on this, but apparently it isn't hard to get suspects to (quasi) confess to crimes they didn't commit once you start providing incentives to come clean. I think unfortunately the best policy is the most draconian--if you have a case where there is little reasonable doubt, you have to fail all parties. Many more complicated schemes setm from admirable motives but carry in their wake bad incentives.

I don't think automatic failure is really an incentive, though, unless they really thought they cheated and thus had reason to think they would be expelled otherwise.

Jeremy,
I wasn't suggesting that your plan was unethical or ethically problematic, only that the one that Mullins describes is because it creates an incentive to q-confess.

But that's what I was responding to. I can't see the motivation to q-confess on that plan. The options for someone who didn't cheat are not to confess and get nothing or to q-confess and fail the course. Who would q-confess in such a situation? The only people who would confess in such a situation are people who believe they cheated and think it's likely that they are the ones who were identified.

Clayton,

I have a hard time imagining that someone would q-confess. It isn't as if people simply come forward and say "I cheated" and that is the end of the matter. People come forward and tell you what they did that constitutes cheating. People that cheated generally know that they were cheating, but there were a couple of cases where people weren't sure if their action counted as cheating. In cases where the action isn't cheating then no foul. I don't really see the problem here.

Jeremy and Matt,

Here's something that has happened to someone I know.

(i) Prof. made Machiavellian proposal. Announces to class that she knows someone cheated and says that if guilty parties do not confess by a certain period, the punishment they'll receive will be more severe than if they do confess.
(ii) During this period, a party who did in fact cheat does confess to cheating.
(iii) Later, the subject recants saying that they feared they knew they were under suspicion and said they confessed for that reason thinking this would be the best way to minimize losses under the unfair and arbitrary rule that is the philosophy classroom.
(iv) Parents who q-know what darling little angels they've raised cite the coerciveness of this incentive scheme in threatening to take the case to court, administration, etc... saying that their darling but timid child who has a long history of being wrongly accussed of crimes was strong armed into confessing to avoid being expelled from the university. Even if you don't get q-confessions, you'll get people who claim to have q-confessed and try to use this against you.

This is a real case that happened to someone near and dear and if it were not for the students inability to keep his story straight when talking to his parents, I have no doubt that there would have been the threat of a lawsuit. It would have been a meritless lawsuit, but that's all it takes for the administration to completely override your judgment, threaten you, etc. Now, I do suspect that the incentive scheme will on occassion elicit q-confessions. For example, you might have students that are paranoid and think you suspect them of crimes they've not committed (this was part of the student's story. The proctor, he said, had it in for him and he cited the proctor's behavior to back him up. The behavior was typical of a proctor who saw someone cheating). At any rate, here is a second kind of concern that arises in these cases. I think it's best just to be straight with the accussed.

Oh, also in college, an entire dorm knew of a case in which two roommates had submitted the same work where one student received an A and another a D. Word got out and the message was this--the prof. grading wasn't a competent or fair grades because [insert explanation here: sexist, racist, drunk, illiterate, etc...]. Not a huge worry but looking back on it, I'd like to think that the prof. was in fact a Machiavellian and not just partial to blondes like Rob.

In my case, all I did was ask students to admit to any greater cooperation than I had allowed. When they came forward, I looked at their exams compare with those of their classmates and determined if there was enough similarity that I would later have questioned. If there was, I treated them as if they had done half of the work that was too similar, and I asked them to do more questions. I made sure I didn't do this with anyone whose questions weren't similar enough that I would have caught it later. I believe that gets around your worry, because anyone having to do more work from coming forward would simply have lost the points otherwise.

Yeah, that sounds much better than the Machiavellian approach. I assume all worked out for the best?

Over drinks a friend suggested this. When you catch a plagiarist, you call him to the front of the class and tell him that you want him to give a presentation. Secretly behind him you have displayed the internet cite from which the paper was pilfered and you highlight as he reads. You thank him and inform him that he's failed the course. When he objects, you ask him to turn around and check out the screen. I suspect that would be a quick ticket to lawschool or wherever it is one goes when one is kicked out of the profession.

They way I see it, if you aren't "in" or tenured there is only one simple rule: Unless you have proof, don't say a word. The tie always goes to the runner. Obvious speculation and obvious feeling around risks alienating your students. They're students, not troops. There are many more exciting and educational ways to be a hard ass.

Often when a prof simply makes it pointedly known that he/she is on the lookout for cheats, it reduces the incidence of cheating.

For obvious reasons, it's tricky to lay the charge unless the proof is irrefutable, as in the case of an essay plagiarized directly from an on-line source.

I recall an incident at University when a prof came close to accusing a friend of mine, A, of cheating because his paper was similar to the paper of another friend, B. In actual fact, A and B often spent long hours in the pub discussing literature, rather than actively helping out with each other's papers.

If only one of them was in your class and had discussed the topic with the other?---- they are guilty of discussion only, apparently .
Such discussion is the essence of Philosophy -- it has a long and honorable history. Judge them only on the quality of the writing and ideas. They have done nothing morally wrong.

Precisely - I think it is unfair to penalize students who may reflect similar or even identical discussion points in papers.

To my way of thinking, cheating involves the use of incorrectly or poorly understood material that has simply been lifted off the internet or copied from a text source. It's fairly easy to detect when a student is writing about matters he/she clearly doesn't comprehend, because there is often a rote quality to the writing and poor analytical skills.

Recently an associate received a paper that had clearly been lifted from the internet. When she Googled the first sentence, she found the source. The student had done little more than copy and paste - without even a cursory attempt to paraphrase.

I agree - debate and discussion is an integral part of the educational process.

I'm having trouble how you can think of it as perfectly moral to engage in the kind of collaboration that I had explicitly told them would count as cheating on this particular exam. If there's nothing wrong with it in principle, it's still wrong if their instructor gives them an explicit prohibition on that kind of discussion. The only thing they were allowed to discuss was where in the text the answer could be found. They were allowed to share the notes they took during class but no other notes. I was quite clear on what they could do, and a number of students came forward to tell me that they did more but thought it was ok.

If you teach for more than a semester, it's extremely unlikely that you haven't had someone handing in internet materials with copy and paste. I had someone do it my first semester, and it happens almost every semester if I do regular papers and not dialogues (my general preference for introductory courses) or flat-out exams (as I did last semester).

I agree that that's much worse cheating, and perhaps I should have allowed them to collaborate more on their take-home exams, but surely what they did was morally wrong given that I had explicitly told them not to do it and that I would count it as cheating. Given that my solution to the problem was just making them do a little more work, I don't think there's a problem with expecting them to do their own work in this case and then counting it cheating when they didn't.

I was speaking in general re connection between debate/discussion and perceptions with respect to cheating.

Yes absolutely, if students are explicitly instructed that they are not to engage in discussion on a given subject area (especially in the context of a specific assignment/project), then it's a different matter altogether.

Just to be clear, in my earlier post I was referring to general discussion of a philosophical/literary nature, not to discussion centering on an exam per se.

A few people on blogs where I have posted have mentioned difficulty linking to my blog. If this url doesn't work try searching - Aidan Maconachy Main Blog

I go along with the notion that its wrong to grade them as if they were cheating simply because the evidence seems rather weak (and I assume some statement like - don't work together - was stated).

Still, if they did work together, its much more likely they did so online. If they had and one found answers similar to those written - then I would suppose cheating (without collusion) for each (thus, even if they were working together, their grade would not be based on that suspicion).

The evidence wasn't weak. It was extremely strong. It just wasn't so strong that it was absolutely impossible that they didn't work together.

There's no way they had simply found a common online source. My questions are very particularized, and the evidence in this case wasn't that they had used some source that would have given them all the same answers. They did even the same questions, when they had been given a large number of questions to pick from. Doing the same questions, making exactly the same points in exactly the same order, and using all the same terms where it counts but different terms where it doesn't strikes me as extremely strong evidence that they worked together in figuring out what to say about the questions they had together chosen to answer, and that is exactly what I had told them would count as cheating on this particular exam.