What should I do as a teaching assistant when I find cheating behavior from actual grading?

Question

Update:

I found 8 students cheating in today's quiz despite stern warning and personal threatening to kick them out of the classroom if I notice cheating behavior again. 4 of them also likely cheated last time. The main reason is I left 5 minutes to fetch old quiz from last time.

I have reported this to the instructor and the two chairs of the department as well as sending the quiz sheets. I am very upset even though it "is not my fault". I have asked the professors explicitly that I do not to see these students in classroom again.

---

I work as a probability&statistics TA. During actual grading of the quiz papers submitted by students, I suspect some students are involved in cheating. The question is what I should do. I should note that my relationship with students is not the best judged from past experience.

The evidence that support my claim students were cheating in the quiz was follows:

1. During the quiz, I noticed students A, B, C, D, E, F, etc sitting very close to each other. I pulled their chairs apart and minutes later their chairs were close again. While I did not suspect cheating, I felt this is quite strange. In the end I have to pull their chairs apart three times. I also noticed they talking to each other, but I am not sure if they were borrowing the calculator or something. I simply did not suspect cheating.

2. During the actual grading I found the students A, B, C, D, E, F, etc all submitted work of very low quality and they made identical elementary mistakes like 1+1=3 on their exam sheets. This elementary mistake was carried through to the second part of the exam, such that a few of them did not bother to give any derivation to the wrong results in the test paper.

I have reported this to the professor, who avoided my email on any discussion with this topic. My questions are as follows:

1. If I want to report cheating, how do I make sure students A, B, C, D, E, F, etc all cheated? Of course there is a small chance that they all made the "stupid mistake" due to some random misunderstanding. For example, maybe students E, F did not cheat; they simply misunderstood the problem or their Casio calculator malfunctioned. Who knows?

For example, when I was an undergraduate, I was wrongly accused for cheating, and I knew such accusations makes people psychologically very uncomfortable even if turned out false in the end. While I am quite confident with what I found, I do not want to be the mean professor who treated me that way. I checked the university honor code and it says "the instructor should communicate with the students regarding the nature of the charge and the evidence...". I simply do not know what to say in this case. Should I simply say "I am suspecting you of cheating behavior, please explain yourself", or something?

2. More importantly, what I should do for the future to the class to prevent cheating? I put "cheating behavior means -10 points and an invitation to visit the Dean" in every exam sheet. But I could not prevent this situation from happening again in future. I felt very uncomfortable that I am preparing lecture for students who paid negative amount of input to the class material. To me a student walking out of the classroom and believe my lecture was tedious is okay; one do not need to take the class if one already knew the material.

But cheating behavior is far worse; it makes the normal Q&A process break down, and I simply do not know what feedback should I give to the students who cheated. It also makes life very unfair for students who made an huge effort but did not do as well as cheaters. I find it very difficult to prepare the lecture in the same mood again and pretend that this have not happened.

3. In the extreme case, if the professor took no action at all, what should I do next? Should I waste many hours coming back and forth on this "trivial" issue and facing various committees, or should I simply turn a deaf ear on it because this is first time offense? To me, this seems a black and white situation. But I am still quite confused.

Answer 1

Addressing (3): You should check the policies at your institution, but a common rule is that the only person who can initiate academic dishonesty proceedings is the instructor of record for the course, ie the professor. This means that you cannot prosecute the students yourself; your job is to report what you observed to the professor, and let her handle it. (Note that references to "instructor" in university policies most likely do not include TAs.)

You say the professor is "avoiding" your emails about this, but how do you know? One possibility is that she hasn't received them, or hasn't read them; so you should contact her by some other means and ask if she got them. Once you have verified that she did, that is where your responsibility ends, unless you are asked to further explain your evidence, appear at a formal hearing, etc.

Another possibility is that she got your report and decided not to take any action. Again, policies vary, but most likely that is her prerogative. Maybe she thinks the evidence is not strong enough, or knows something about the disciplinary procedure that you do not, or simply wants to avoid conflict. Whatever. Her problem, not yours.

A third option is that she is taking steps to punish the students, but has not told you about it. Again depending on policies, she may not be allowed to do so; such proceedings are usually confidential, and you don't really have a need to know.

So in short: make your report to the professor, verify that it was received, and move on with your life.

Answer 2

Disclaimers: Firstly, how cheating is treated varies a lot among different countries, so what I'm going to say might not be applicable to your case. Secondly, you actually ask several questions, and I shall concentrate just on the following one:

More importantly, what I should do for the future to the class to prevent cheating?

The students' behaviour you describe in point 1) should be cut off directly during the exam, and not discussed afterwards. In general, I enforce the following rules:

1. Students are not allowed to talk to each other. If I find two or more students trying to talk to each other, I warn them once, but the second time they're out.
2. Students should bring their own calculators and cannot borrow calculators from other students. If a students forgets her/his calculator, too bad, she/he can employ the good ol' pencil and paper method of calculation.
3. We don't have movable chairs, but, in case: if you're too close, the first time I come to pull you apart and warn you; the second time, that's the door and you're welcome to walk out of it.

Answer 3

If I were on an academic misconduct panel hearing your charge, I would not give much credence to subjective reports of strangeness and sitting close, unless the distances were strikingly abnormal. In light of the physical constraints of the room and the typical spread of students in that space, and especially assuming that these students were close friends a.k.a. study-buddies, how far off of the norm was their spacing? [Suggestion for the future: require cheating suspects to sit in completely different areas during exams; put it on the syllabus that during exams, you will randomly move students within the room, at your discretion]. Your second basis is, potentially, clear and objective, and is the evidence that these panels require.

I am a bit confused about your second basis, that they all claimed that 1+1=3 -- that is so incredibly unlikely that I have to believe that you are speaking somewhat metaphorically. If that is really the case, then that's a key to your dilemma. Suppose, for example, that the accused all made the same elementary mistake in confusing "standard deviation" and "standard error". In a sense that's a very elementary mistake, but in another sense, it's a beginner's mistake that's likely to be propagated in an under-informed study group. However, if you are talking about a really elementary arithmetic error that is exactly repeated in the group, and you were not speaking metaphorically, that is an event of such low probability of occurrence that you could probably say why it can't be due to a non-cheating common cause. So it is crucial to carefully examine the nature of the shared error.

Typically, the instructor of record has to be the one to make the formal accusation. You should request a meeting with the professor (yes, not pleasant) to discuss the problem in person (forget email), and be prepared to suggest a three-way meeting with the chair. If pursuing the matter would not cause you severe problems (the professor hating you forever and voting to terminate your funding because you stirred the pot), I would say you should pursue the matter. If there would be negative consequences for pursuing justice, it comes down to the question of how much you care about justice. [A totally neutral way of framing the matter, of course].

Answer 4

This is an answer to your second question: "More importantly, what I should do for the future to the class to prevent cheating?"

I usually give different sets of exams; usually three is enough. Make sure students sitting next to each other have different sets assigned to them. Have the students sign an attendance sheet indicating what set they were given. (This sheet can be distributed while they are taking the exam. While they are filling it up, verify that the set they indicated in the sheet matches the actual set given to them.)

Let's say the sets have the following questions:

Set A: 2+5=, 6+2=, 3+3=, 0+9= (answers: 7 8 6 9)

Set B: 2+4=, 7+2=, 4+4=, 0+7= (answers: 6 9 8 7)

Set C: 3+5=, 5+2=, 5+5=, 0+6= (answers: 8 7 10 6)

Let's say that students X, Y, and Z were given sets A, B, and C, respectively. If student Y is seated in between X and Z and answers 7 8 6 9, then this is extremely strong evidence that student Y copied from student X (and not from Z). (The example I've given has four answers copied, but if there are much more, then the evidence would be stronger.) Keep the test papers of students X and Y (and maybe even Z) as evidence, but do return copies of the test papers to them.

Show the evidence to students X and Y. They may or they may not admit guilt. If they do, then obtain proof of it (a signed statement or a video). Submit all the evidence (including the admission of guilt) to the student disciplinary body*. There is a very good chance they will find the students guilty.

By showing the students that you report cases of cheating and that you win them, students will be less likely to cheat in your class in the future.

Note:

*I realize that my answer here is more appropriate for teachers of a course, and not for teaching assistants. If, in your institution, a teaching assistant is not supposed to go directly to the student disciplinary body, then present the evidence to the teacher. Your problem is now whether or not the teacher will act on it.