75

So a TA came to me asking for advice. He gave wrong information in office hours leading a student to a wrong answer.

The TA showed the student how to solve the problem via the wrong statistical test - the assumptions for the statistical test he used do not hold in this case. The answer generated from this test yields a test statistic that doesn't make any sense when you think about it's meaning. Kind of like if you were asked to "find the length of the fence that encloses ..." and your method produced the answer of -50 meters. The TA provided the student with step by step instructions, but did not do any of the numerical calculations. Had he done those calculations he would almost certainly have caught his mistake.

The student complains that she should not be punished because her TA made a mistake. TAs can change HW grades, if they have good reason to do so, and this TA wants to correct his mistake by rewarding the student the points on the problem. I told the TA that I would get back to him as I don't want to give my advice in haste. I think I want to say something like this.

to the TA: "Thank you for asking for my advice in this situation; it is indeed a difficult position to be in. However, this situation concerns me, not because you gave wrong info or bad advice, as we all occasionally make mistakes, but because you seem to have told a student the method for answering the question. While it's OK for students to get help from their TAs, it should only be by reviewing concepts and asking them 'leading questions' that nudge them in the right direction. Note I am not saying that what you did is objectively wrong, as I should have been more clear in my instructions at the beginning of the course, but in the future telling students how to do a problem, in my opinion, is unfair to the students who can't make your office hours."

to the student: "Your TA contacted me recently about changing your HW grade. While I sympathize with your situation, and understand that it is very frustrating to receive bad advice/information from your TA, it is ultimately your responsibility to make sure your solution is reasonable, and therefore I can not give you full credit on this problem. Ultimately the majority of the points you lost were not for using this incorrect method (the TAs error), it was for not interpreting your answer and realizing it made no sense. As stated on the first day of class and in the syllabus I require all students to interpret their answers to make sure they make sense. Please come see me if you would like advice on how to check your answers as this is a very important skill in all scientific disciplines and in life in general. TAs, solutions manuals, and even professors make mistakes. Never trust something just because it was told to you by someone with more authority."

Is this the right thing to do?

14
  • 75
    I would like to speak for the student. IMHO, you're too harsh to him. The reason he is in your class is to learn. If he knew how to make sure your solution is correct, he would be in a more advanced class.
    – Nobody
    Mar 22, 2014 at 5:29
  • 27
    (I also am inclined to think that not giving points back to the student in this situation is standing on a valid principle but is almost inevitably going to annoy the student to the extent that it will interfere with (i) their performance in the course and (ii) their opinion of you. Is the principle worth that much?) Mar 22, 2014 at 5:57
  • 10
    I meant the student's solution. In other words, he is in your class to learn how to check the solution. I was quoting the words you wrote. Sorry about the confusion. At any rate, I think it's your and the TA's responsibility to help him to learn. What I would do is to explain it in the next class to all the students, not just that one, so that everybody, the students, the TA and even yourself learn it as a lesson. Just my 2 cents worth.
    – Nobody
    Mar 22, 2014 at 5:59
  • 25
    "Never trust something just because it was told to you by someone with more authority." Sorry, this is weak. The TA is your assistant. The TA is your representative. If the TA is flawed, that's not the student's fault; that's your fault for not vetting a better TA. Mar 22, 2014 at 11:47
  • 30
    Further, the very essence of your teaching position consists of people "trusting" because you're someone with "authority". Sure, learners should check their work to understand and verify, but it's unrealistic to expect that students should raise suspicious eyebrows to each and every word that's taught until they can check it and verify for themselves. Not all knowledge is as simple to verify as an algebraic equation. As a general principle, yes, it's good to have a questioning mind and not take everything at mere face value. But realistically, it's unrealistic to expect such as an absolute. Mar 22, 2014 at 11:51

11 Answers 11

126

I agree that in principle, the student is responsible for critically evaluating information from all sources and rejecting any that's incorrect. I wouldn't simply award the student points for the incorrect solution, but on the other hand it seems unhelpful to just mark it wrong and leave it there (and despite your explanation, the student will likely perceive this as unfair). In similar situations, I've done one of two things:

  • Allow the student to revise the answer, and regrade the problem; or

  • Discard the problem for that student, and recompute the score for the assignment based only on the remaining problems.

5
  • 9
    +1 For the first bullet point. Just ask the student to redo that problem and to ask you any questions about it (so he gets the "correct" help if he requires it) and then you grade the resubmission. As to the second bullet, I'd say the student should do the problem because the student does need to know how to do the problem and HW is a good practice/feedback mechanism. Mar 22, 2014 at 9:25
  • 3
    If you discard the erroneous problem, what would you do to the ones that figured out the correct solution despite the incorrect instruction? Discount their work as well? Award them bonus mark? They possibly had spent extra hours trying to research the problem themselves that wouldn't have been necessary if they had received correct instructions in the first place.
    – Lie Ryan
    Mar 23, 2014 at 16:55
  • 3
    @LieRyan: I'd give them full credit for a correct colution and apologize to them for the confusion, but I don't think it would be appropriate to award them additional points in compensation for their time. Grades are supposed to measure the quality of a student's work, not serve as an arbitrary reward/punishment mechanism. If I felt that in figuring out the issues involved, they had really shown exceptional initiative and understanding, I might consider bonus points. Mar 23, 2014 at 17:14
  • 4
    No other student saw the false instruction in this case. Hopefully if many students saw this false instruction (like in a TA section) than most likely the error would have been pointed out by one of the good students, and likely a very informative discussion would ensue. I'm going to have the TAs go over this problem in section, presenting the wrong solution and then asking the class why it's wrong and truly making this a learning experience for all. I've been told by the other 20 TAs that many students did exactly what this TA did. Mar 23, 2014 at 21:22
  • I generally like the idea of disregard a problem when the reasonableness of the problem for measuring competence is called into question.
    – Joshua
    Sep 24, 2017 at 18:55
51

At the moment I'm not sufficiently clear on what the TA did and why it was wrong: in plenty of contexts "[telling] a student the method for answering the question" is exactly what TAs are supposed to do. But again, I'm not sure exactly what happened nor what your understanding with the TA was.

So let me only address the issue of what to say to the student. The principle that they are ultimately trying to get at what is actually correct, not what any one other person tells them is correct, is certainly a key one in a STEM field. In fact, to me this is one of the distinct and pleasant features of mathematics in particular: my freshmen students are very chary to point out even the most trivial errors in my boardwork, but they will eventually learn that if I forgot to multiply by 4-x then (i) my answer is wrong, (ii) it is partially their responsibility to point out that my answer is wrong (the consequences of it being wrong are going to hurt them more than it hurts me, since they are trying to learn how to do problems that they haven't done before), and (iii) if they do point out that my answer is wrong they can be confident that I will react to it constructively rather than defensively.

Students in STEM fields need to learn how to correct the errors that they encounter from teachers and texts, and they do: story about how I mangled something in my graduate course this morning and the students got me to see that what I had done was bogus omitted (okay, drastically abridged). But lower level students have so much trouble with this: errors which to us look insignificant ("of course it should say this...") can really ruin their day. They're trying so hard to get a clue, and many of them are working in a state of cognitive dissonance that comes from having 99% of their inherent intellectual curiosity beaten out of them in their pre-university education. In other words, they have unfortunately been highly trained not to try to understand but rather to do what they're told. Most university students do clearly understand that more is expected of them and, thank goodness, many of them eventually regain the intellectual curiosity they lost, but you can't be an effective university instructor without knowing that many many students are just going to swallow anything you say which is too complicated for them to chew through in five seconds.

So when an instructor of any kind tells students something wrong, it is a big problem. (Wrong is really the nightmare limiting case; I spend so much time trying to avoid correct explanations that I have learned from experience they will interpret incorrectly.) If I say something in a lecture and realize later that it's wrong, I will come back and fix it in the next lecture. If I have something which is wrong on study materials, I will try to correct it and send out a new copy ASAP. If I tell them something that's wrong while they're studying for an exam, I won't wait until the next lecture but send along an email owning up to and fixing the mistake.

This is pretty much my point: students should -- ideally, and eventually -- take responsibility for correcting mistakes, sure, but so should we, and the two are not mutually exclusive. Giving students points back when we make a mistake does not undermine the lesson of their own responsibility, in my opinion: it just does needed damage control. When I ask a false question on a midterm or final exam (and it does happen, unfortunately) I either throw it out from the grading or give everyone full credit. I am not testing them on their ability to correct my mistakes: to do doesn't seem instructive to me but obnoxious. I remember once that a PhD student whom I was preliminarily advising took a qualifying exam which had a whopper of a false problem on it. As soon as I saw the problem my reaction was "I will be shocked if that turns out to be true." But I have more experience than the students and the luxury of having that reaction: my being wrong about it has no significant negative consequences for me. It turned out that the student had received less than full credit for his solution to the false problem! My reaction to that was essentially "WTF? Please grade this exam again." And they did.

I would strongly advise you to give the student the points back on the homework assignment. (Added: or discount the assignment, or let them redo the problem, or whatever so long as they are not getting irrevocably dinged for the TA's mistake.) And sure, do make it a teachable moment -- talk to the student about how they can learn to better adapt to the very distressing situation of being given wrong information. Tell them that at some point they will acquire enough expertise to look at an "authoritative document" and say "I will be shocked if that turns out to be true." But don't allow the TA's mistake to lower their course grade. They'll never forgive you for that, and they won't learn how to correct their mistakes in the future.

8
  • 3
    @MHH: Thanks, I did eventually see the additional information you provided. And yes, making sure that whatever you do will not give an unfair advantage to the student is certainly a good idea (although similarly to Arrow's Impossibility Theorem, for years I have dimly suspected that given all the "axioms of fairness" I feel like I should apply to grades, I could logically deduce that I must give every student an A). But really in many courses one problem on one homework will count negligibly to the final grade. The effect on the student's psyche may not be negligible, though. Mar 22, 2014 at 7:54
  • 1
    Another thought I had, by the way, but managed not to put in my answer: "If I were your TA, I would have kept this to myself: better that the instructor knows about neither the mistake nor the remedy, perhaps. Better for the TA, anyway." Mar 22, 2014 at 7:58
  • 12
    Having said that: I do get frustrated as an instructor when students go elsewhere for help with their homework and get less than fully competent help, which they then try to take out on me. For some reason this hasn't happened for a little while, but for years I used to get complaints like "Even my tutor couldn't do this homework problem." It made me so frustrated: I don't know who the hell your tutor is; I certainly didn't vet his CV or anything like that. Why are you drawing any conclusion from this other than that you may want to find a different tutor?!? Mar 22, 2014 at 8:03
  • 2
    As an instructor, I wouldn't draw a strong distinction between myself and the TA. The TA is part of a team that the instructor leads, and so the instructor should own TA mistakes as if they were his own. That may bear upon what sort of response makes sense. Mar 23, 2014 at 7:21
  • 3
    In other words, they have unfortunately been highly trained not to try to understand but rather to do what they're told. This is a very good point :) Mar 24, 2014 at 9:54
16

While I sympathize with your situation, and understand that it is very frustrating to receive bad advice/information from your TA, it is ultimately your responsibility to make sure your solution is correct, and therefore I can not give you back the points.

How is a student supposed to make sure their solution is correct, if not by seeking help from external sources such as TAs? I don't think your logic holds up here, and for that reason, it's not reasonable to penalize the student for the TA's mistake.

On the other hand, I'm not suggesting giving points for incorrect work. Some "neutral" solution, like omitting that problem from the computation of the student's grade or allowing them to redo the work, seems appropriate.

Please come see me if you would like advice on how to check your answers as this is a very important skill in all scientific disciplines and in life in general. TAs, solutions manuals, and even professors often make mistakes. Never trust something just because it was told to you by someone with more authority.

I don't think it's really practical to tell your students never to trust something just because it was told to them by someone with authority. Of course they should try not to do so more than necessary, and hopefully "necessary" means less and less over time, but some amount of trusting people with authority is necessary to jump-start the learning process.

It may be useful to make the point that even people with authority - professors, TAs, etc. - are beholden to a higher authority, namely correctness and self-consistency. If a student thinks something seems inconsistent or wrong, they should look into it further, by doing their own derivations and/or comparing multiple sources, until (hopefully) they eventually come to understand the thing that didn't make sense in the first place. If someone has made a mistake, it will be revealed in this process.

4
  • 3
    "I don't think it's fair to tell your students never to trust something told to them by someone with authority." I think you left out the key word in the passage you quoted: just. Mar 23, 2014 at 8:51
  • If your teacher told you that the optimal fence like to enclose a pin was negative one hundred meters, you would trust him? This was a case where simply interpreting the answer would have led to suspicion. Mar 23, 2014 at 21:25
  • 7
    Sure, if you're the faculty member who knows the subject. The student is probably very confused, that's why they 1) sought out help, like every faculty member tells them to, and 2) trusted that help. If they understood the material they wouldn't have done (1), so they wouldn't have had to evaluate it or not.
    – chmullig
    Mar 23, 2014 at 23:36
  • @MarkMeckes fair point, I think I found a way to edit it.
    – David Z
    Mar 31, 2014 at 1:13
11

I agree with your council to the TA but you should give the student the points (or not use this assignment toward the calculation of the student's grades).

Is it fair that the student is receiving credit for incorrect work? No, but the point of any learning environment is to promote learning and I think your draconian adherence to grade fairness in this case would undermine that environment.

This student appears to have made a good faith effort to learn, and the TA appears to have made a good faith effort to assist learning. Exercising some compassion will ensure that this continues.

7

If you were the one that taught the material incorrectly, would you still hold the student accountable for your mistake?

The TA is your assistant. You must take responsibility for what the TA is teaching. The TA's words are your words by proxy. However you resolve the situation, you should address it as if you personally made the error.

Perhaps the error was so obvious that you'd expect the student to catch it even if it came from you, and finalizing the grade is reasonable. The fact that your TA made the error instead of you is immaterial to that decision.

5

A student cannot be expected to trump their instructor's judgment when they know who grades the answer. What if the student said, "yeah I had my doubts as well, that's why I spoke with my TA?"

Although I think you should give the student points or a redo in any event, I think you should also consider whether it was realistic for the student to be able to point out his TA's mistakes? Would most the other students awarded points be able to do this?

Often in statistics the mathematical derivations require a much deeper understanding than simply learning to use the tools. True in many disciplines in fact. It sounds like you're holding this student to a much higher standard, which is why I think you should reconsider your position.

4
  • So if you were told by your instructor that an optimal pot of coffee contained negative 30 grams of ground coffee beans, you'd not try and figure out if the instructor made a mistake? This particular example didn't require deep intuition; It required thinking about the problem rather than blindly copying a recipe from a cook book. I have told my class up front that this will be different than the standard intro stats class and that I will expect them to think and interpret data (within reason). This has surprisingly led to great teaching reviews (I thought there would be more resistance) Mar 23, 2014 at 21:32
  • That said I am glad I posted this question, because I do agree that this time, punishing the student via their score is a bit too harsh. Thanks for your feedback. Mar 23, 2014 at 21:33
  • If you think his/her error demonstrated a lack of understanding that other students demonstrably posses, then it's harder to fault your stance. But did the question allow the other students to demonstrate such proficiency, or to borrow the cooking book analogy, did they get points even if they just followed the recipe?
    – NameSpace
    Mar 23, 2014 at 21:59
  • Oh yes this student and others that followed this cook book method were rewarded partial credit. The student wants full credit. Mar 23, 2014 at 22:19
3

I think the best thing to do is let the student re-do the task, and have a talk with your TA. Making a mistake is human, but if he worked out the whole problem using the wrong method - without realising the result didn't make sense - I'd say that's an issue.

You expect your students to interpret the results of the calculation, and you expect them to choose the right method of solving it on an exam, so when a TA makes such a mistake, I'd question him.

I'm not saying that making a mistake is the problem, it's the nature of the mistake. When I lecture on programming, I sometimes make "thinko's" or get a design element wrong, but I won't make a whole algorithm in a faulty matter. It's a bit hard to compare lecturing programming to the more mathematical courses, because blackboard mistakes in a calculation are easily made and overlooked.

Yet, I have never used the wrong throughout a whole problem, I'd realise I'm using the wrong method as one point or another, surely when the result is wrong (Once again, easier to notice in programming), but you'd expect a TA to be able to interpret the result, no?

Ofcourse, I have no idea what the actual mistake was he made, but from your post it seemed like a key mistake.

3

From my experience, homework exercise are there to help students learn, and are graded fairly lightly. The fact that your TA showed your student how to solve a problem, might or might no be a problem depending on the circumstances. For example, if a student had showed effort and reasonable understanding of the problem, and the TA explained the problem so that the student understood. No Problem. Of course if your TA just did a problem of the student, then there is a problem.

How difficult was the problem that you TA explained wrong? The fact that your TA doesn't understand the material, at least not all of it is a problem, in my view. Does your TA come to your lectures? Your TA should understand the material well, and they should be able to explain it. It is his job to help the students, not confuse them.

I would explain the right method to the student and give them full credit for the problem, and have a talk with the TA to make sure that they understand the material.

2
  • the TA made a reasonable mistake; the problem arose because he only told the student what to do, like a step by step recipe without asking the student questions to see if she was following (a full solution minus numerical calculations). He did not get to see the numerical answer that resulted. If he saw that he would have realized his mistake (since the answer that method would give is nonsense). After seeing the answer I am sure he would have been able to redo the problem correctly. I think this was just an honest mistake by the TA. Mar 22, 2014 at 14:53
  • 1
    OK, I still think the fact that the TA gave a wrong method is a problem, perhaps he should be more careful. You could advise him to due the assignment themselves first. As for the student, if the answer obviously made no sense, like negative variance, I would probably still give her full credit as the mistake was on our part, but advise her to be more careful next time.
    – Akavall
    Mar 22, 2014 at 17:19
3

I worked as a tutor the last few years of my undergraduate studies and I wanted to add that to a student seeking help, a TA (or tutor) is someone you believe. Your student would not think that what the TA said was wrong because they are a TA. If the TA made an honest mistake, as a teacher, you should bring both the TA and student together and explain what was wrong. You could then give the student half credit if they answered in the way that the TA erroneously showed them. You could use this experience to help the rest of the class. I do not profess to understand STEM, as I have a degree and History and am working on my MA in education, but I understand teachable moments. If you want to prevent this in the future, meet with your TA and make certain that they are helping students using the correct process.

1
  • This is very specific to STEM. In history it would require an unreasonable amount of work for a student to verify everything an instructor or TA said. In some STEM classes we cover about 5% of the material covered in a history course. A class may only go through 70 pgs of material total over the whole semester! We focus on understanding and independently verifying nearly all of it. Answers to questions should at a minimum always be checked by interpreting the results. The student failed to do this. They were not punished for the error, they lost points for not checking their answer. Apr 10, 2014 at 2:42
0

The notion about the "fairness" of reliance on TAs for advice being in question because not all students may be able to attend a TA's office hours is an utterly false premise. Instead of being "equal protection" it's "equal jeopardy." Bad idea.

If a student is so brilliant that they need no instruction to get correct answers, or if you're such a brilliant lecturer that you reach every single student with the exact best way to deliver the material for 100 percent comprehension...then eliminate TA/student interaction entirely. However, in my experience, such students and such teachers together are countable on one hand in a given university, usually with fingers left over for other tasks.

If the student came up with the answer that she should have, using the incorrect method, then she should receive credit...if she redoes the solution using an appropriate method, which in your grand mangnaminity you should bestow upon her. And yes, she should be chagrined to have not noticed the nonsensical answer.

0

Invalidate the question and apply again, if possible, renormalize the grade if not, to the whole class, to be fair...

Expanding:

Sorry to be blunt, but, as others said, you made a mess. Any attempt to rationalize it would be pointless, just take the hit (remember: your TA... and everybody makes mistakes, how do you deal with them that's important).

If you just award the points, you risk other students coming out of the woodwork saying the same thing or worse, because it is not really a fair option, neither is not awarding the points. Since both options are unfair, redo the whole deal.

My point is that this part of the assignment/exam is now tainted. Any result from it, should be removed. For the whole class. Drastic, I agree, but, IMHO, the best way to deal with it.

If you still have time/availability, re-explain the topic to the whole class, making perfectly clear the right way, the wrong way and why the wrong way doesn't work and re-apply the assignment/exam. This way you even expand what they learn by showing one possible pitfall...

If you can't, remove that item from the grades and scale the value of the other items to fill the void.

Either way, make everything public and clear to everyone. Sometimes even the smell of "cover up" is worse than the mistake.

1
  • Can you please expand a bit?
    – jakebeal
    Sep 23, 2015 at 17:09

You must log in to answer this question.

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