Our books

Become a Fan

« We're hiring! | Main | What can be done to improve the philosophy grad school experience? »



Feed You can follow this conversation by subscribing to the comment feed for this post.

Sara L. Uckelman

In the UK, borders for individual classes are sharply fixed, and cannot be changed. Borderline cases are resolved holistically. If a student is within .5% (or maybe 1%, I don't remember) of a border, then the board of examiners has the option of bumping them up to the higher grade class. This can be done on the basis of two criteria:

1. Preponderance: Are the majority of their individual class marks in the higher grade category? Maybe they had one really bad class that they bombed, but did really well in everything else.

2. Trajectory: Has their mark gone consistently up over the course of all three years? If you take their course work in the final year alone, would they be in the higher category? If they wrote a thesis, was it in the higher category?

These criteria can be individually sufficient, but sometimes jointly necessary.

Michel X.

What if a student has 3.5 borderline points? =p

Is there any harm, really, in bumping *every* borderline grade up? One could do this by singling out all of the students on the cusp of a higher letter grade and bumping up that way, or even just by saying that at the end of the semester any partial credit gets bumped to a full percentage point (so, e.g., 89.3% turns into 90%), or anything above .5 (normal rounding rules apply), etc.

That would be my inclination, at any rate. It's easy and seems perfectly fair to everyone involved.


I don't announce point thresholds in advance. That lets me avoid borderline cases. I'll explain.

I say on the syllabus how much of the final grade will be determined by each component of the class. At the end of term I come up with percentages for all components of the class. Then I look for gaps. If, e.g., there is nobody between 89 and 91, then that's a good spot for a gap between B+ and A-.

This is a form of curving, which I'm fine with. Of course it's not guaranteed to work--if there are lots of students and the grades follow a non-gappy normal distribution, then it won't work. But given the numbers of grades I usually have to come up with (typically in the 10-20 range per class), that doesn't usually happen.

Trevor Hedberg

Michel X, I have known professors who handle it that way, but I worry that strategy just pushes the issue back a bit. If you round all scores, then you run into a similar problem with a student who makes something near an 89.4%: how confident can we really be that this student didn't make the magic 89.5% threshold and shouldn't get the bump up? Part of the reason I like my system is that the students have control of their fates in these circumstances, and it doesn't come down solely to my fine-grained judgments on a few assessments.

Michel X.


If you round all partial percentages up to the next whole number, however, then the 89.4 becomes a 90. Seems fairest and easiest to me, and nobody gets partial percentages any more.

Jerry Green

Good post, Trevor, and good discussion.

I have an explicit policy in my syllabus, as follows: "The letter grade cut off for, e.g. an A- is 90.0, not 89.5 or 89.9. But I may choose to round up in exceptional cases, if (i) I feel you’ve done better or worked harder than your grade suggests, and (ii) your grade is not due to excessive missing assignments. This is a courtesy, not an entitlement."

As you can see, it basically just codifies the problem you're asking about. Technically, cut-offs are strict and well-defined, and there's no obligation on my end to consider rounding-up in borderline cases. But I may do so in certain cases.

In practice, I'm much less strict than the policy makes it seem. This is mainly because, as you note, grading is highly fallible. So, in practice, I think its safe to assume that I could have given them an extra 0.n points somewhere during the term. And the way I run my classes, students have to really try not get the grade they want, so its usually pretty clear who deserves rounding up and who doesn't.

One point about effort: you're right that its hard to discern how much effort a given student is actually putting in, let alone how much that effort should determine their grade (We wouldn't dock points from students who don't have to study for lack of effort, after all). I set up my classes explicitly so that you can get an A just by working hard consistently: there's basically a threshold above which high quality work doesn't make your grade better. At least, that's what I do for lower-division courses. Upper-division/ grad courses are different, but in lots of ways that make rounding-up less of an issue anyway (at least in my experience).

Jerry Green

One more thought, about avoiding the problem rather than addressing it. As you mention, putting the students in control is good for both student and instructor.

I often will allow for extra credit assignments that put the rounding-up burden on the student rather than on me. I think they're a lot like your Borderline Points system, except I just treat them as extra assignments. For example, I'll let them do a small number of one-page response papers or something, worth 1% of their final grade each (so, much more work for these points than normal). Anyone who is in range of getting rounded up can choose to do some more work in get over the hump, if they want. But its on them.

I also like to allow re-do's on certain assignments. For instance, if I assign three short papers, I'll let them R&R them if they get a C or worse, with a new grade of up to 80%.(And I do mean R&R: they have to write a cover-sheet explaining what they fixed and why). Its a very powerful teaching tool, and it also lets them do the work to 'get rounded up' themselves.

Third, for lower-division courses I have small, low-stakes quizzes every day. They're worth about 1% each, but I'll offer more than I grade; this semester, I give 38 quizzes, but you can only get 35 points total. This build-in curve allows students to miss a few quizzes' worth of questions throughout the term without penalty. This means that they're less likely to end up with just a point or so less than they needed, and if they do they have a clear explanation of why (e.g. they blew off 5 quizzes instead of 3).

One principle underlying all these approaches is to give the students several opportunities to get the grade they want, and if possible more than one avenue for doing so. Done right, you minimize cases where one assignment is directly responsible for the grade. And putting the students in charge of their own grade as much as you can makes your life a lot easier, not to mention making the students happier.

Trevor Hedberg

Jerry, I also incorporate rewrites and exam retakes into my courses, but I hadn't made this connection to how it could impact the frequency of borderline cases and how students feel about them. It sounds like our overall outlooks on these issues are quite similar.


If a student has a course grade of 90.1%, is that student questionable to deserve an A? Why should a student with two points less out of a 1000 points course with a course grade 89.9% be considered for less or scrutinized MORE? It's simple math. If your syllabus says that 90% is an A, math tells us that anything from 89.50 to 90.49 is considered 90% (using mathematical round up/round down principals). Just give the student that best rewards the student and don't look to punish border, because that is totally counterproductive to encourage students to work hard and learn. And some of the hardest working students are those on the border that will do anything to get those those 5 or 10 points that make the differ. Use some common sense and don't judge students so harshly.

Verify your Comment

Previewing your Comment

This is only a preview. Your comment has not yet been posted.

Your comment could not be posted. Error type:
Your comment has been saved. Comments are moderated and will not appear until approved by the author. Post another comment

The letters and numbers you entered did not match the image. Please try again.

As a final step before posting your comment, enter the letters and numbers you see in the image below. This prevents automated programs from posting comments.

Having trouble reading this image? View an alternate.


Post a comment

Comments are moderated, and will not appear until the author has approved them.

Your Information

(Name and email address are required. Email address will not be displayed with the comment.)

Current Job-Market Discussion Thread

Job ads crowdsourcing thread

Philosophers in Industry Directory

Open thread on hiring timelines