The Harvard Crimson headline “70% of Faculty Vote to Overhaul Harvard Grading With A Cap” sums it up: Harvard University faculty voted, beginning in Fall 2027, to cap the number of students receiving an A in courses at about 20% (with some wiggle room). To go along with that, GPA will be replaced by percentile rankings in various internal awards. A survey of students suggests near-total opposition, with 85% of students disapproving.
(Ignoring that everyone should stop talking about Ivies all time whenever we discussion higher education in the USA,) There’s some things about this that I like. The idea that a B or C might as well be failing is a harmful expectation, and the expectation that everyone can or should get an A is a norm I’d like to see changed. Grade inflation is genuinely happening across much of American higher education, including apparently at Harvard. There was a report last year that most grades at Harvard were As, and it seems unlikely that the average Harvard student has gotten so much better in the last couple of decades as to explain the trend solely through academic improvement. It would be good if the expectation becomes that getting an A is unusual or exceptional, that there’s nothing wrong with not getting an A, and that students with a B or C average will probably be fine after graduation.
This isn’t a perfect implementation, though, and I’m not sure I’d want my employer to introduce a similar policy. Especially with smaller classes, sometimes you just get better groups of students. I suppose the argument is that a class where more than one in five is A material is too unlikely to be a concern. That makes sense for 300-person classes, but I’m not sure about classes of 30 or 20 or 10, especially upper-level classes in difficult fields which have already done a fair bit of weeding out weaker students just to get there. And for at least some classes in disciplines with more objective measures (like mathematics), exams set to a common standard should be able to create a fair grade distribution that doesn’t require any kind of percentile ranking.
This decision raises a bigger, more fundamental question. What’s the point of grades? Grades are supposed to evaluate students, but the kind of grade inflation common in so many schools seems to me to indicate that they aren’t terribly useful for that purpose. Grades are motivation for students in only the worst way, external validation replacing intrinsic motivation. They are a blunt tool for measuring student learning, and other forms of feedback are much more useful.
It’s not obvious what grades are even supposed to be doing, when we use terms like “evaluate” or “measure.” Are grades a certification that students have met a certain standard? Are they feedback for students, for parents, for employers or application committees? For my history classes, which aren’t particularly unusual that I can tell, a student’s final grade reflects a fuzzy combination of student effort and improvement, soft skills like turning work in on time or attending class, and actual content mastery. (And no matter what people say, it’s very rare that someone genuinely thinks a final grade should reflect content expertise–you don’t think there should be late penalties?) I’m not quite a grade abolitionist or radical ungrader, but it’s hard to imagine that we would create anything like the same grading system for high school and undergraduate education that we have today, if starting from scratch.
There would be some slight difficulties, especially for external purposes (“List your GPA”-type applications), but I would mostly do away with letter grades, if I were in charge. Classes instead should be graded Unsatisfactory (earning no credit), Satisfactory, and Excellent, with a soft cap on the number of Excellent students for each class (perhaps the faculty member would have to write an explanation of giving more than 5% of a class/students taught in a semester a grade of Excellent, or having an objective exam/set of exams with an Excellent cut-off of a score that historically only 5% or fewer of students get). At smaller institutions, the expectation would be that most students will receive a Satisfactory and that more classes would have no grades of Excellent than would have multiple ones.
For most people, what’s the purpose of going to college? It’s to get a degree, which serves as testimony of being educated, in the sense that you prove you can follow directions, show up at least some of the time, not prove to be so incompetent or objectionable that you get kicked out of school, and the like. If we think about a college degree in the most mercenary sense, entirely in terms of preparation for future employment, then it really isn’t necessary to differentiate between A, B, C, and D if prove you can get through college effectively (or at least consistently above a failing level) for most people. Will some people put in less effort or produce lower quality work than they might under a traditional grading system? They might, but lots of people already put in less effort or produce lower work than they’re capable of doing.
From my fluffier the-humanities-really-matter perspective, starting class with the expectation that most people are going to pass, we can instead concentrate on actually learning instead of trying to quantify everyone’s contributions on an arbitrary scale. Students can replace a focus on grades with a focus on learning.