Next semester, I am going to take a course in which I have some background knowledge. Given the relevance of the course in my research, the lack of available students with sufficient knowledge in the course (it is one that is rarely taught at this school, and most of my supervisor's students either graduated last semester, or have just started with him this semester), and my general need to find some courses to TA for, the professor (my supervisor) had offered me the grader position for the course. The professor would be grading my own assignment, and I would then be marking other people's assignments based on those. It appears this situation has occured and been asked about before on this stack exchange, but not from the point of view of the TA. I initially expressed my discomfort with this situation, but, upon being told (by my professor) that this had been done before, decided to accept, to maintain a good relationship with my supervisor.

I however have a number of issues which, as the beginning of next semester, are bothering me more and more:

I know a number of the other students in the class, and we often used to organise study groups and work on the assignments together. Would it be ethical of me to still participate in these study groups ? On one hand, I will not have access to the solutions until we submit our assignment, on the other, I will definitely be able to see if someone makes an obvious mistake in their proof that I, as a TA, would deduct points on.

I was initially thinking that the best course of action for me would be to cut off all relations related to the class (that is, remain friends with the students I know, but not bring up the class with them), but at the same time, I would thus be depriving them of help that I would have given to them as TA had I not been in such a situation. I'm finding it extremely difficult to set the line between "We are fellow students together and helping each other out" and "I am your TA and you are a student, and the help I am providing here is so you can properly learn the class".

Updates in responses to questions:

  • I have asked the professor for clarification; he says this is not the first time he has done this
  • I could certainly ask for "blind" grading
  • I am taking the course as a student, but have some background in the course
  • Comments are not for extended discussion; this conversation has been moved to chat. OP provided some responses-to-questions in the comments; these have been edited into the post.
    – cag51
    Commented Dec 22, 2019 at 3:40

11 Answers 11


Actually, I think the best course of action would be for you to decline this offer. You have different personal relationships with different people in the class and it will be difficult to put aside all of your unconscious feelings and grade in a completely unbiased manner. It also opens the possibility of your friends putting pressure on you, which you can resist, but which could become uncomfortable.

An exception would be for grading that is purely objective, not requiring any analysis or judgement of the quality of the answers, such as mechanically recording the answers in multiple choice questions.

However, if blind grading could be arranged so that you grade paper not knowing who wrote them, then my reservations would probably disappear.

Blind grading would leave you free to interact normally with study groups and would lessen or remove any pressure that could be applied against you.

At a minimum bring up such concerns with the professor and point out the uncomfortable position it puts you in.

Perhaps there is a possibility that you help out the prof by grading a different class and suggesting you grade this one in a future offering.

  • 28
    The notion of "blind grading", especially with regard to close friends, is difficult due to recognizing handwriting, I think. Commented Dec 18, 2019 at 23:11
  • 4
    @paulgarrett. True enough. Unless the answers are typed.
    – Buffy
    Commented Dec 18, 2019 at 23:12
  • 19
    @Buffy even typed, in some cases. After 3-4 years, we started recognizing people's coding style for instance. Nothing is ever really blind (like when you review papers in a narrow field, it's blind, but everybody knows the authors and sometimes the reviewers)... Commented Dec 19, 2019 at 14:58
  • 3
    Of course, if it's mechanical grading, then it wouldn't be necessary to have someone with knowledge of the course and anyone could do it, not just the OP. Commented Dec 19, 2019 at 15:48
  • 9
    When I had to grade my friends, I would grade their work last, so that I could see how I graded the same mistakes before. (How many minor errors did I find? Where they big enough to count as a -1 or a -2? etc).
    – nomen
    Commented Dec 19, 2019 at 17:09

Just to go against the grain, if you decide to take it:

Be methodical and very objective. Every action you take should be documented and based on provable facts.

I taught friends for more than a few courses. I usually do a pass cataloging errors, then I assign a mark for each error, which leads to the grades. That way, when someone questions the markings (Brazilian national sport at universities btw), you can literally point at the rules and how the grade was constructed. A friend that was letting me crash in his couch for the duration failed one of my courses and we are still friends...

For the day to day stuff, since you don't have access to anything confidential, just be a better than average student that helps people, whenever you feel like doing it, but at least during your TA hours. Be careful with the ethical lines (like if your professor mentions something that is not public knowledge), but don't overthink it.

  • 7
    I assure you that this is a sport not only in Brazil. Some things are fairly universal :) Commented Dec 19, 2019 at 23:42
  • 1
    At least in CS in my experience it's completely impossible to come up with an objective, consistent and complete way to "catalogue errors". This might work somewhere where there's just a single way to arrive at a solution and you just have to look up where someone went off the rails. I also can't imagine how this would work for say History essays.
    – Voo
    Commented Dec 20, 2019 at 11:11
  • 2
    @Voo well, I did it in CS (and intro to statistics). The idea there was checkpoints.Then I would categorize how they deviated between each checkpoint. Break those ways into 3-4 "badness" classes, take 25% of the marks for each. The main idea is you don't associate the error with the person. You associate the mark with the error, then map it back. Commented Dec 21, 2019 at 0:36
  • @Voo I had some later courses where the professor had written up a set of API tests for our assignments and ran them against your program and each test that failed was a mark off the score. Of course, that meant that if your program didn't follow the specification, you got a zero mark, but then that's the real world, too. Commented Jan 12 at 18:56

I have some experience with this kind of situation. Two or three times in my career as a professor a student was in a situation similar to yours: too much knowledge to take the course but for various good reasons needed to. I arranged for them to (help me) prepare solutions to hand out and sometimes grade papers.

Everyone in the class and in the department knew about this arrangement. It never seemed to cause any problem. My general attitude toward homework may have helped ease any potential friction - the purpose of the assignments was to encourage learning, not to assess. So they were commented on extensively, marked just [excellent|done|please see me for help] and did not count heavily toward the final grade.

Your situation may vary.

  • 2
    +1 for the general course philosophy that helps sidestep the problem.
    – Buffy
    Commented Dec 19, 2019 at 15:54
  • 4
    Which country is that? You have a university system that is not broken. Commented Dec 19, 2019 at 17:45
  • 2
    @CaptainEmacs UMass Boston. I think that at many places there's enough professorial independence in the classroom to allow for this style. Commented Dec 19, 2019 at 18:13

I would politely decline the appointment. I would be uncomfortable with the conflict of interest and, though certainly flattered, uncomfortable with the ethics of and my qualifications to grade a class I was still taking.


I have been a TA and grader for a course that had my friends (and, in fact, my girlfriend) in it. (I wasn't also taking the course at the time, but that doesn't seem like the main issue here.) Each grader had a "conflicts list", and nobody graded work by anybody on their list. (Being mere acquaintances wasn't a conflict, but being close friends or romantic partners was.)

I think being the only grader for this class would be untenable in your position, but if the professor grades not only your work, but also your friends' work, then it doesn't seem quite as fraught.


I think it is fine if you accept your professor’s offer to be a TA, though I would let them know your discomfort about the situation. I think the best course of action would be to hold normal office hours as a TA so that you are not depriving your fellow students of help while also maintaining professional boundaries. And as others have suggested, blind grading seems to be the most fair way of assessing classwork.


I was a TA/grader for courses that had friends in it. I graded impartially. If they got the question wrong, they were marked down appropriately.

I know a number of the other students in the class, and we often used to organise study groups and work on the assignments together. Would it be ethical of me to still participate in these study groups ?

Why wouldn't it be? As you point out, you won't have the answer key in front of you.

I will definitely be able to see if someone makes an obvious mistake in their proof that I, as a TA, would deduct points on.

I don't understand how your position as TA affects this. If you weren't the grader, you would point out the mistake and help your friends fill in the gaps in their knowledge. Because that is the point of a study group. You're also taking for granted that you will know all the answers. (And if you're the TA, don't you do "office hours" to give students help anyway?)

Honestly, if I needed the job, I would take it but it's a little funny of your prof to have offered it in the first place.

  • Yup. I was going to say this exactly. I always graded my friends last, so that I could see how I graded other people for the same mistakes.
    – nomen
    Commented Dec 19, 2019 at 17:06
  • Also, good point about the study group. You're a student until you have the stack of homework assignments in front of you.
    – nomen
    Commented Dec 19, 2019 at 17:12

Apart from the ethics, let it be known that, you may not teach (or grade) a course which you do not have appropriate documented knowledge of. If you must take the course yourself, you obviously have not. This clearly indicates that your knowledge in the field has not been documented, yet. If you already have an appropriate higher degree and taking the course is simply for the paperwork, this practice may pass. But as a normal student, grading direct peers is actually forbidden by accreditation policies. They say "teaching personel must have appropriate qualifications". At least in Germany (this question has no country tag).

  • That is in theory. I have seen this exact same thing being done in Germany. Of course, the student assistant did not sign any grade sheets, but then again, not even the TAs/PhD students can sign grade sheets...
    – wimi
    Commented Dec 20, 2019 at 12:13

You say:

I know a number of the other students in the class, and we often used to organise study groups and work on the assignments together. Would it be ethical of me to still participate in these study groups?


was initially thinking that the best course of action for me would be to cut off all relations related to the class (that is, remain friends with the students I know, but not bring up the class with them)

I think you should not take this position for two reasons:

1) It almost unavoidably creates "the appearance of a conflict of interest", which is as bad as an actual conflict of interest.

2) It puts you outside of your own community. This is unhealthy for your own social/academic well-being. As a student it's not good to get cut off from your peer group. You draw support from them, discuss tricky assignments with them, commiserate with them on setbacks. These are important things for anyone, students no exception. But to fulfill this position in a fair way cuts you off from uncomplicated interaction with your class.

Basically, taking this job puts you in a very bad position.


Here is another possible solution, depending on your situation: you can TA without officially taking the course, and enroll in something like a reading course instead.

As an undergrad, I needed to learn complex analysis, and I wanted to take the graduate course as I was also trying to get a Master's a the same time. There was no graduate course that semester, but the professor teaching the undergraduate course (which I hadn't taken) proposed the following:

  1. I would officially sign up for the graduate course with him, which we did as a reading course. (I believe I was officially enrolled in the graduate complex analysis course, so that is what showed up on my transcript.)

  2. I would attend and TA for his undergrad class (grading + office hours---in fact, I ended up giving a couple of lectures when he was traveling).

I think this worked out well with no conflict of interest as I was not a student in the class, and everyone just thought of me as the TA, but I got the benefit of the lectures and doing the undergrad homework before grading it. In fact grading the students' homework and answering their questions helped me learn the material better.

FYI, I don't recall having any issues with grading the undergrad problems. I did separate exercises for my reading course, but I'm sure if I had trouble with any of the problems I was supposed to grade I could have consulted with the professor---as should be the case for any course you TA.


If you are uncomfortable, don't do it. It's one thing if you WANT to do it and it has been done before. But if you don't want to, just...don't. Let your prof do a little more work. Don't do something like this just to butter him up. You even have the perfect excuse.

You must log in to answer this question.

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