Harvard capping high marks; should we?

[poetryandwine]

Harvard University academics have voted to cap the percentage of ‘A’ grades awarded in undergraduate course modules to 20% of enrolled students, plus or minus four students. There is no cap on other grades, including the grade A- ( ‘A minus’). This will be from Autumn 2027. The change will be assessed after three years.

About 70% voted in favour of the cap.

The grade of A at Harvard is supposed to be reserved for work of exceptional merit. The last time the percentage of A grades was as low as 20% was 2005. In 2024-25, about 60% of awarded module grades were A’s.

Students are not happy. There is a fundamental difference between what students want from assessment and what academics and employers want from assessment.

UK First Class degree awards have undergone a similar but less dramatic change over the last few decades, especially since tuition fees increased. We know employers don’t find undergraduate marks and degree classifications as useful as they used to, and why they now set such high bars.

Do MumsNetters think that capping the percentage of First Class Marks in each undergraduate course module, which can easily be achieved by rescaling marks similarly to what we already do, would be a good idea, or not?

Have the standards decreased, or have the students gotten better, or something combination of the two?

“We are moving the oral exams throughout as a result which will no doubt disadvantage ND and possibly international students but it's hard to know what to do.”

If the goal for all is long term employment, then oracy and help there is surely what is most required with gentle adjustments from an early age, for ND (and international students).
Isn’t that the whole point? They need to develop the skills required for further down the line.

@Mithral Yes, she’s 10 years “call”
but the attributes are still more than academics. Maybe week long internships? Take a look at possibles?

I think oral exams are used elsewhere. DD did them in an Italian university.

@poetryandwine
I think that 80-90% of students now get a 1st or a 2.1 at university. IMO that is way too high. It makes it impossible for recruiters to distinguish between candidates.
As it is easier to get a 1st or a 2.1 at a lower-ranked university than a higher -ranked one, the degree classification then becomes almost meaningless especially with blind university recruitment. I think this has led to employers using multiple tests themselves to distinguish between candidates.
Students with very mediocre grades at A-level are getting 1sts at lower-ranked universities. So I think would it be unfair for students at higher-ranked universities with all A star grades at A-level to have the number of 1sts capped when their contemporaries at lower-ranked universities are not.

I think it worth considering why degree classification inflation has occurred in the UK context. Prospective students are looking for a degree to support their career aspirations. Many things feed into this including subject, uni standing/reputation and degree classification. Uni administrations are well aware of this and their is no incentive to be more harsh/stringent with classifications. Even without uni blind recruitment prospective students are not going to favour a uni (that is similarly placed in the league tables) that awards substantially less 2:1s and 1sts than its competitors. Students achieving higher classifications is supported by many factors outlined by @poetryandwine They include more formulaic exam content, more formative support, reduction of unseen controlled assessments etc. On the surface things may seem similar to previously, but the support and information prior to the assessment may be markedly different. This is all without considering the can of worms that is AI.
There are checks and balances including external examiners and professional body accreditation, but the system is much less standardised than GCSEs, A levels etc. In a world where the vast majority of graduates get a 2:1 or a 1st, being a uni that awards 30% 2:2s isnt going to be a selling point. It is a market and the students and prospective students are the customers.

I am genuinely interested in unis using oral exams. How are reasonable adjustments implemented for students with anxiety or ASD that impacts their oral communication. Are they permitted alternatives?

Not being snarky - a genuune question.

@Owlbookend You could also argue that employers are the customers of the grads and the universities. They don’t feel they get what it says on the tin.

Yes - i agree in one sense. But it is the students who pay the fees not the employers. It isnt a situation i would design, but it is the one we have.

@Owlbookend Lots of them don’t pay though do they? Parents do if they earn more of course. However the bursaries are there for students from low earning households plus full loan. They are taking a gamble on job success or they don’t intend to pay it off by doing very little.

The growth of universities shows this model doesn’t hold back applicants . Vast numbers never pay the loan back so we, the nation, pay. Lots of students just think they do.

Theres also the big question about why are dc going to university if its not to get a job? It’s clearly not to make coffee. So why did they go if it wasn’t to get a job? They are customers but they need to understand employers are the customers at the end of the chain. Many them are looking at employees from abroad.

@Owlbookend
I think long-term universities will lose out on customers if they don't start having some integrity in the way they award 1st and 2.1s. If degrees become meaningless and students cannot get jobs after a degree, they will vote with their feet.

My perception is that the UK university grading system has very little in common with the US system. The UK system probably requires reform but the US system has little to teach it.

I am fascinated by the difference in essay quality on my DC's course between a low 2:1 and a First - it's massive! Why does so much hang on the difference between 63% and 73%? The grading system ought to be using a far greater range of marks. My DC, right at the top of the cohort, has had marks between 63% and 84% over the course of 3 years, with an average around 75%. Why aren't top students getting marks of over 90%? Grading is incredibly culturally loaded.

As I said, this isnt the situation I would design it is just the reality. The high proportion of 2:1s and 1sts is pretty much sector wide. A UK university could step out from the pack and engineer a lower proportion of high classifications similar to Harvard. They wouldn't need to do it by curve based grading, there are plenty of mechanisms that could produce this result. However, for the reasons outlined above I think such a move is unlikely at least in the short term.

Due to the market based system universities have to attract students. Funding is based on student numbers. Regardless of whether tuition funding comes directly from.students' families or from a goverment backed loan they want to attract the funding by maintaining recruitment numbers. They know that a significantly lower level of 2:1s and 1sts than competitor institions is likely to impact applicant numbers for multiple reasons. Nothing is stopping them taking a different approach, but in the current context they don't.

@fairyring25 And many young people should, except they cannot get decent apprenticeships. Look at the Blair company issues. They don’t have good enough alternative options.

The universities were allowed to lift the cap on student numbers in 2013. Oxford didn’t raise their intake. Not sure about other elite unis but the others went he’ll bent into huge recruitment and building mode racking up debts. They didn’t have to do this. Some could have stayed smaller and more true to themselves. It’s pure greed and unfettered expansion. And who actually benefitted? Difficult to say in 2026.

Universities need to satisfy their customers, the students. These YP have become very good at learning within tight bounds, so tight that the border with memorisation is unclear. Within these bounds, they are being taught more effectively from a young age right through university.

When assessments are very rigid, almost to a template, most students who work diligently can do very well. But very small deviations in exam content and structures are throwing them.

In my STEM field, a problem written as a whole is more difficult than one broken out into several steps. It can also be more satisfying, as a student can often find an alternative solution to the one suggested by the multiple steps. This type of problem is seldom set anymore. Only very strong students can cope; the others are thrown and often really shaken by an exam they perceive they cannot do ( even if that perception is wrong, even if marks will be scaled, etc).

I think the change is that top marks, in exams, are obtained for doing routine, much practised things very well. They used to mean more.

It looks to me like this starts back at GCSEs and certainly A levels, where I sometimes see the Physics and Maths problems that generate a lot of complaints. The bar for these, intellectually, seems to be to be declining. Again, a problem designed to stretch pupils or differentiate between them is called unfair.

This mindset has come to university. I cannot believe it is useful in the workplace.

So true. Nothing will change in the UK unless joint action is taken.

Many students hate oral exams. For some good reasons (the unpredictability, the feeling of exposure, the adversarial setup) and some bad ones (worries about examiner subjectivity and the inability to cheat).

Of course it is worse for some SN students.

Also, they are very labour intensive.

For all of these reasons I don’t think using oral exams at scale is likely to catch on.

PS @Owlbookend We do require an oral exam of every project or dissertation student. My only ND project student was fine with this and did well.

We had discussed that the point of the exam was to demonstrate that he’d done the work, which he knew I believed he had. And he had reason to think it was good. I emphasised he would need to do this in the future, as he was aiming for a PhD, so this would be good practice. And that by preparing well he could consider it an opportunity to show us his stuff.

Quite different from an exam in a course unit, however.

I agree that it seems incompatible with blind applications.

But in (parts of?) the Civil Service at least, blind applications have had the effect of increasing the proportion of Oxbridge intake.
I wonder whether the trend will continue?

@poetryandwine I found your post on the fact that students can't creatively problem solve whole problems in STEM at university level fascinating and worrying. I believe the UK was renowned for creative problem solving in science before so we shouldn't lose it.
IMO, GCSE and A-level is being taught to the test due to the publication of detailed mark schemes. This needs to be stopped. However, top universities have the ability to make their exams anyway they like. Why don't they make students do tests with difficult problems that only the best can solve. If students aren't challenged, they won't become good problem solvers.

Who indeed benefitted?

Not the YP I don’t think.

"In my STEM field, a problem written as a whole is more difficult than one broken out into several steps. It can also be more satisfying, as a student can often find an alternative solution to the one suggested by the multiple steps. This type of problem is seldom set anymore."

This seems extremely disappointing. Do you get the impression that this is also the case at COWI universities (assuming the subject in question is either math or physics)? Or maybe even Oxbridge? Do you think there needs to be a return of AEA in physics / a greater use of the math AEA, or perhaps giving the ESAT / TMUA some written problems rather than MCQs?

I think it is worth noting a couple of things about institution blind applications. First, they arent usually qualification blind. So employers can see if a student has gained As at A level etc. Second they will often involve assessments of reasoning and/or technical and professional skills. If a student has a strong academic background and has further developed at a highly selective uni they will likely do well in the process. It isnt 'ahh they got a first give them the job'. Students at less selective institutions are not dismissed based on uni reputation, but to suceed they will have to demonstrate competence in a range of challenging recruutment activities.
There are few signs thst highly qualified applicants are abandoning more selective institutions in order to get an 'easy' first at a less selective uni.

This is an interesting perspective. In the context of blind applications for employment, it certainly matters. If employers were known to value high quality degree programmes would it matter less?

Agree.