Is it acceptable, if too many students in the class would fail otherwise, to award additional points across the board (in the one class, in one semester)?
Would it require approval of the Division Dean? How could this be justified?

  • 4
    This question is begging an entirely different question: why are so many students failing? Is the course too hard? Is there something wrong with instruction? What would cause such a thing to happen? Dealing with the problem this semester is a bandaid if this happens repeatedly.
    – Compass
    Oct 16, 2014 at 18:54
  • 1
    I agree with you in that there is a bigger question at play with a situation such as this; however, my question stands. I have knowledge of this occurring and am wondering if others have encountered a similar situation.
    – J Anderson
    Oct 16, 2014 at 19:11
  • 3
    So... no curves are allowed because it keeps lots of students from failing, but this year the result is that lots of students are failing? What changed?
    – jakebeal
    Oct 16, 2014 at 19:59
  • 4
    I don't think we can answer this question. In most cases (at least in the US), the instructor of a course has wide latitude to determine how letter grades are assigned, consistent with the syllabus as distributed to students, and general principles of fairness. If, as it seems, higher authority has restricted your latitude in this regard, then you will have to talk to that authority to determine what latitude you do have. Maybe you have to talk to your dean, maybe not. It completely depends on your internal policies and politics. Oct 16, 2014 at 20:36
  • 2
    I agree with others that more info is needed. On what basis was it determined that "too many" people are failing? I know of many cases where, say, a professor created a midterm exam, many students failed, and on reviewing the exam the professor then realized that it was indeed too hard. In such a situation, various sorts of curving or point-adding may take place, and no one sees a problem with that. There's a difference between A) the students really did not learn the material as required; and B) they did, but the mapping of their learning to "points" was faulty.
    – BrenBarn
    Oct 17, 2014 at 6:07

1 Answer 1


I would say, in nearly all cases*, it's unethical. Specifically, if a student's submitted work does not fulfill the requirements for a passing grade, then they do not deserve a passing grade. Arbitrarily adding points that do not reflect the quality of submitted work makes the grade partially or completely useless as an indicator of performance in the class, and if it's not good for that, what's the point?

However, unethical does not necessarily mean unacceptable. Don't get me wrong, I think it should, but I'm not going to impose my moral standards on your grading scheme. Ultimately nobody on this site can tell you whether it's acceptable to do this. Your superiors at your educational institution might be able to, since they'll at least know whether it would violate any of the institution's rules.

In any case, as a few people have mentioned in the comments, arbitrarily adding points to keep students from failing is a band-aid effect. The fact that too many students are failing indicates that there is some deeper problem. Perhaps your teaching is inadequate. Perhaps your grading standards are too harsh. Perhaps the students are not properly prepared for the class. Perhaps your idea of what constitutes "too many students failing" should be revised. And so on.

*Speaking of harsh grading standards, this is (the?) one case in which I think it is ethical to add points: to compensate for an overly difficult exam or assignment. This is a little tricky to get right, though. You have to have a preexisting standard to measure whether the exam is at an appropriate difficulty level. For example, in classes I've TA'd for, there has been a longstanding policy that any time the average grade on a test (across ~1000 students) is less than 70%, it will be taken as an indicator that the exam was too hard, and therefore enough points will be added to everyone's exam grade to bump the average up to 70%. Given the large class size, I think this is reasonable.

For smaller classes, you have to consider the possibility that you just have an underachieving group of students one particular year. In another (small) class that I took, each year the instructor would compute the distribution of all grades in the class over the previous ten years, compare that to a desired target distribution, and make the course material for the next year harder or easier as needed. From what he told me, it gave pretty consistent results. I think this is a good way to do it.

The key is that you have to have a preexisting criterion for when points will be added. If you look at the grades after the fact and decide, hey, too many students are failing, let me add some points so we all look better - then you're doing it wrong (IMO).

  • +1 for this. I'll add a situation where I think changing scores after grading is acceptable. We once had a Final where most students did well or ok on most problems but almost nobody solved problem 7, but some people went off in a totally wrong direction. The TAs who hadn't seen the question before also thought it strangely worded. Because of this, the Prof. decided to turn it into a bonus question. The grades looked much better after that but very few people went from failing to passing. Yes, he did take the previous year's distribution into account but that wasn't his reason for the change. Oct 17, 2014 at 9:37
  • I think the first paragraph is strongly worded, although the later paragraphs soften it. I would usually view "the requirements for a passing grade" as certain levels of mastery of the course material. Grades measure this level of mastery only indirectly, and with a significant amount of measurement error. Numerical grades can give a false sense of objectivity and precision to the grading process, especially in the American A/B/C/D/F system. Oct 17, 2014 at 12:02
  • @OswaldVeblen (late response) indeed, it was my intent to use strong words. I think they are called for here. Also, of course in an ideal world a grade measures mastery of the material, but in practice we usually set up some objective criteria that we take to be a proxy for mastery of the material. I take the latter view of grades in my answer, i.e. the pragmatic one where the grade is based on the quality of submitted work, and I choose to ignore the fact that this is an imperfect test of mastery because I think that fact is not relevant to my answer.
    – David Z
    Oct 25, 2014 at 22:41

You must log in to answer this question.

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