I noticed an error in a graded exam during office hours. Should I give the student the lower grade?

Question

You are a professor. A student comes to you with a question on an exam they have taken and received back. The student wrote 5 as the answer to a question, and they want to defend why 5 is a correct answer and they should receive full marks. Their argument is invalid, but you notice that the answer was misread as 9 and the student was given partial credit (5 is entirely incorrect and they should have received zero credit). Do you regrade the question, giving the student zero points, or do you leave it alone?

Arguments for regrading the question:

• It would be unfair to the other students to give this student marks they did not earn.
• Pursuing accurate grading is what is important and is the objective of the course.

Arguments for leaving the question alone:

• The student came forward with a good-faith intention to learn why their argument wasn't correct, and shouldn't be penalized - this would break an implicit contract, and encourage students to not seek further feedback.
• The unfairness actually stems from the misgrade itself, and there isn't anything wrong with leaving the misgrade alone now that it already exists. After all, if the student hadn't come forward, the misgrade would still have been left alone.

Answer 1

Let the grade stand, but point out your error.

The principle I try to follow is that students shouldn't be negatively affected by faculty mistakes (or that such effects should be minimized).

I wouldn't suggest adding an additional assignment to maintain the grade, since this is effectively another kind of penalty. But I would suggest being transparent with the student, explaining your mistake, and showing them where they went wrong (otherwise they are deprived of the feedback that the correct grade would have provided).

If the exam were graded on a curve and re-grading this one question on Student A's test would materially improve other students' grades, I would give other students the boost (i.e. the grade they would have received had Student A's exam been correctly marked the first time) but leave Student A's grade intact.

Answer 2

You should never penalize a student for taking the time to come to office hours (or whatever) and seeking additional help or clarification. Not only would it make it less likely for that particular student to ever ask you for that type of help again, but it could very well disincentivize other students as well.

Answer 3

To me, it is actually odd to see all answers but one say that you should leave the points as is. I do agree however that this situation should be handled somewhere in the policy, shared with the students at the beginning of the course, and this particular instance handled leniently if no such policy exists.

When I studied, it was (implicitly) understood for all courses that when asking for a re-grade of a question, it will be examined with additional scrutiny which could result in the final marks for that task to go either up or down. Several of our instructors repeated this explicitly as well.

However, I can see how different approaches might work better for different cultures. In my country, we typically call after-exam consultations "complaints" (even though there is a second valid word translating to "insight" or "inspection"). The majority of the students came in to see if they can "find" an extra point or two, and very rarely to actually understand where they made mistakes.

I don't think there was anything inherently wrong with such a system, as it was a good fit for the culture: the policy was there to discourage "begging" for points, but there wasn't any negative consequences to asking the instructor to point out where you went wrong in your partial answer to question 5A. Additionally, for explaining whole concepts that the students did not understand, they were always welcome to come to regular consultations (not related to an exam), and ask away whatever they have trouble with.

Answer 4

For many of the reasons you state, I would let the original grade stand. I viewed my job as one of teaching, not grading. Overall, the change in the grade will probably be a minor thing, but the effect on the student's psyche should be considered.

However, you can also take advantage of the situation to create a teaching moment. It may be too late for this already, and I'm not sure if I'd have been thinking fast enough at the time, but I would suggest making a deal with the student:

You know, I should really change this mark to a zero here, but I'll make you a deal. If, over the next week, you write me a one page summary of the problem and the correct solution to it, I'll let the grade stand as is.

This changes the equation. The student gets a bit of hope. And maybe a bit of respect for you and your willingness to "go the extra mile" to help them. It also, sends them off to do a bit of research that can only benefit them.

---

I'll add that as a normal part of my teaching, I always let students redo work that they want to improve their grade on. Not so much on exams, but in general assignments they could always earn back part of the lost marks by providing a better solution. I would permit this even after discussing the question in class. The educational philosophy behind it is that if you "engage" with the work you will more likely learn it. Just getting marked down for errors doesn't let you integrate the learning. No student ever suggested that this was a form of "punishment".

Answer 5

The key to answering this question is what the syllabus says.

The syllabus is like a contract between the professor and the class describing what happens in different situations that may occur during a term. If the syllabus or departmental guidelines say how to handle the situation, then it makes things easy. (e.g. if you ask for a re-grade in one question, the entire exam is checked)

If the syllabus doesn't handle this case and you want to handle it in a non-standard way, make sure it is handled in the future, and then I would advise handling this situation in favor of the student this time. (That is, don't take off additional points this time.)

Note that this does not mean that every professor should explicitly spell out every possible situation in detail. It means establishing principles so students know what to expect from reading the syllabus, and if you want to do something differently, you then need to spell it out. For example:

• If you want to limit how long students can come back asking for exams or homework to be re-examined, it should be stated clearly.
• If you have a unique policy on how questions are re-graded on an exam, it should be stated clearly.

But, if you have no such limitations, you are always willing to make decisions interpreting rules in the best light with favor to the students, or you put appropriate instructions on each assignment, then this is unnecessary.

Ultimately you should not make a habit of contradicting your syllabus, so that should be the first point of reference in answering such questions when they arise.

Answer 6

You could consider this not just from an ethical standpoint, but also from a statistical one. I remember I overheard a podcast where the commentator couldn't understand "how could correcting an error in data ever be wrong, if you know it's an error?"

They gave the example of looking at temperature data, seeing something was off, finding a 37 entered and referencing a video of the thermometer and seeing it was meant to be 73 (so clearly someone had mistranscribed the temperature entering it in, flipping around the digits on accident). The podcaster meant this as an example of something that "surely couldn't be wrong to correct", and implored (rhetorically) "how could it be wrong to change that back to 73?"

You may not know, but that actually would be wrong. And the reason is roughly as follows: you expect errors like transpositions to be essentially random; sometimes they'll write a 37 for a 73, sometimes they'll write a 97 for a 79. But the correcting of those errors could be systematic, e.g. someone noticing the temperatures seem too low and referencing some unusually low temperatures against the videotape. This bias in error correction would introduce an overall bias in the data.

In the case of correcting errors in exams after they show up to office hour, the bias that is introduced will only harm the honest students, and won't genuinely do anything to make overall the grading more accurate.

If you are concerned with improving the accuracy of your grading, do so by changing the procedures you use for grading. Don't do so at office hours.

Answer 7

You should clearly state the policy in the syllabus and verbally on the first day of class, and when you hand back exams. A reasonable policy is that re-grades may well result in a lower grade. In my 7+ years of college teaching, I didn't do that. I would do so today.

I found students to be very persistent about getting that extra point, in spite of the fact that it won't help their final grade. Since others got two points off for the same mistake, I wanted to stand my ground. Basically, the student was telling me I was wrong for taking off two points instead of one. I was not very pleased with that assertion since it's my decision to make. I know the goals I want to achieve by taking off more points for particular mistakes, especially later in the term.

It is unethical to give a few persistent students a higher grade, just as a reward for being persistent.

Answer 8

Let the higher grade stand but point out that it should have been given no points. You do not want to punish the student for coming for help but also want to let them know that what they answered should be given no marks so that they expect it in the future.

Answer 9

When marking student work I have always told them that if they believe I have made a mistake in the marking then they should identify the error and then discuss it with me.

If, after working through the question, it is shown that I have made an error and they should have received a higher grade then their grade is adjusted.

If it is found that I made an error and they should have received a lower grade, their current grade stands and is not revised downwards.

I believe that generally this encourages them to review their work in comparison to what they have been told and leads to better understanding of the material. If they find or believe they have found an error they should not be afraid that a risk exists of receiving a lower grade by doing so - in effect this would be that there is a risk of punishment for coming forward.

Furthermore, if in reviewing the work together it is found that I have made an error and given a grade that is too high, that is also a learning opportunity.

The actual desired end result of a course is not the actual grade - although that is the metric used for academic success - but knowledge and competence.

Answer 10

It is totally fine to give the student the lower grade once you find that thee initial grade was a mistake IF AND ONLY IF this policy was clearly stated beforehand and publically (meaning all the students of this class have been warned). In fact, I would recommend doing this since it would help to prevent students going to you to try to increase their grades just because they have nothing to lose (except their time) and, on most cases, they would only go when they have some confidence they have a good justification.

Even when the student is "demanding" a better grade (and recognizing it is very tempting to reduce their grade once you find out the first grade was mistakenly high), it would not be a behavior expected from a professor, so would advise against it.

Answer 11

I agree with the general consensus that it is best not to penalise the error in this case. My view is that it is best not to create a disincentive for students to come to you to review their marks. For this reason, I make it a practice in my courses to explicitly set this as a marking rule in the course outline. That way the students are put on notice of my marking practice. For whatever it is worth, here is the blurb I use in my course outlines:

Review of Marks and Consistency of Marks

The course lecturer will make every effort to mark all assessment items in a manner that is consistent from student to student. If students believe that there is an inconsistency between the marking of their own work and the marking of another student’s work (e.g., they get different marks for an answer that is substantively the same) then they should raise this with the course lecturer for review.

The only exception to this consistency principle is this: if the course lecturer finds that a student has accidentally been awarded a higher mark than should have applied for a question, or overall, during subsequent inquiry on the matter by students, the mark will not be reduced, except on request from the student. This is done in order to avoid creating any disincentive for students to discuss marking issues with the course lecturer (i.e., students do not have to worry about raising issues of consistency and then being marked down as a result).