[OC] Top grades keep rising at US and UK schools while adult literacy and numeracy test scores fall across the OECD


oscarleo0

25 comments
  1. Data sources:

    – Harvard: Office of Undergraduate Education grading reports ([https://oue.fas.harvard.edu](https://oue.fas.harvard.edu))

    – Yale: Ray Fair’s grade reports ([https://fairmodel.econ.yale.edu/yalegrds](https://fairmodel.econ.yale.edu/yalegrds))

    – US high school: ACT’s 2022 grade inflation study, Figure 3 ([https://files.eric.ed.gov/fulltext/ED621326.pdf](https://files.eric.ed.gov/fulltext/ED621326.pdf)); grades self-reported, series ends with the class of 2021

    – UK first-class degrees: HESA ([https://www.hesa.ac.uk/data-and-analysis/students/outcomes](https://www.hesa.ac.uk/data-and-analysis/students/outcomes))

    – UK A-levels: JCQ ([https://www.jcq.org.uk](https://www.jcq.org.uk))

    – Adult skills scatter: OECD Survey of Adult Skills (PIAAC), official change values between the 2012 and 2023 cycles ([https://www.oecd.org/en/about/programmes/piaac.html](https://www.oecd.org/en/about/programmes/piaac.html)); Poland’s fall carries an OECD data-quality caveat

    – PISA figure in the notes: OECD PISA 2022 results

    Tools: Matplotlib

    Newsletter article: [https://oscarleo.substack.com/p/grades-go-up-skills-go-down](https://oscarleo.substack.com/p/grades-go-up-skills-go-down)

    Full article for Reddit:

    Over the last two decades, top grades have become the default rather than a distinction. At Harvard, A grades went from a quarter of all course grades in 2006 to six in ten today. At Yale, four of five grades are now A or A-. Among American high schoolers who take the ACT, a majority now report an A average. And in the UK, the share of graduates receiving first-class honors has more than doubled since 2010

    At the same time, numerical and literacy skills are declining. Between the OECD’s two rounds of adult skills testing, in 2012 and 2023, literacy scores fell in 20 of 26 countries. Only Finland and Denmark improved significantly on both metrics. At the same time, in the United States, the share of adults reading at or below the lowest proficiency level grew from 19 to 28 percent in just six years. Also, the average PISA maths score dropped from 489 to 472 between 2018 and 2022, the largest decrease in the test’s history.

    The American high school data best describe the paradox. ACT researchers tracked 4 million test-takers from 2010 to 2021. The share of people reporting an average grade of A rose from 40 to 55 percent. Over the same years, their average score on the actual test fell from 21.4 to 20.2. Same students, better grades, worse results.

    During the pandemic, grade inflation rose to an unprecedented level as exams were canceled and teachers assigned grades themselves. That’s clear from UK numbers, where A grades remained flat since exams are managed by regulators, and exploded from 26 to 46 percent as exams disappeared because of COVID. These numbers dropped again when the pandemic ended, and exams resumed. First-class degrees, which universities award themselves, only stopped climbing when the regulator publicly leaned on them, and are not back to pre-pandemic levels.

    In 2026, Harvard approved a new grading policy to combat grade inflation. From 2027, in any course, at most 20% of students, plus 4 more, can receive a flat A. Ahead of this new policy, the flat-A share fell from 60 to 53 percent during the fall semester of 2025 because faculty were urged to apply it early. Whether that holds will be visible in next year’s data.

    Artificial intelligence is another relevant factor in grades going forward. Among UK undergraduates, the share using AI for assessed work jumped from 53 to 88 percent in a single year. If universities return to handwritten exams, today’s students may struggle after years of other examination formats. But perhaps that’s the best approach to turn the tide on negative trends in PISA and PIAAC.

  2. I can speak from experience and say that UK university degrees outside of the top unis are handed out like participation trophies, Im surprised they hold any academic weight at all by now

    It wouldnt surprise me if A Levels were the same

  3. What is that line????

    “The owners of this country … don’t want well-informed, well-educated people capable of critical thinking … That’s against their interest … They want obedient workers … people who are just smart enough to run the machines and do the paperwork, and just dumb enough to passively accept all these increasingly shittier jobs” – Some old dead guy..

    Welcome to 2026 enjoy your stay.. In a few yrs Idiocracy will be in full effect..

  4. Here in the UK, changes in exams/curriculum in the second half of the 2010s have made education much more rigorous for the brightest students. For those with the potential to achieve a first, their education has probably improved (despite the experience of some other students worsening).

  5. Interesting that the skill decline is not universal across countries. I honestly did not know that, and I would have expected smartphones to have brought skills down everywhere. 

    This would seem to indicate that the problem lies with the US school system in particular. 

  6. There is no link between the two graphs here. The literacy rate decline don’t reflect the same population as those going to elite universities.

  7. Interesting data

    I think it‘s because having good grades gets more and more important and only being mediocre is just not enough nowadays.

    We are more and more focused on telling and showing why we‘re good at everything and better than most.

    Beside the fact that there is also an increase of disparity between people educated, getting a good education and not getting educated at all.
    Some years ago it was important that your children get educated better than you, so they have a better life than you.
    Now it shifts more towards already educated families supporting their children (who can do it better because of average better jobs, more money and having a better understanding of education related issues as well, like how to study) and non educated families staying there because they don‘t understand the need for it, especially with these educated smart asses being all know it all, everytime.

    Furthermore, the school system gets drained as much as possible, so it just comes down to your parents and not the system and the children intelligence (which is quite bad).

    Those are just my 2 cents 🙂

  8. Shit and this is from 2023, before AI. The next results will be catastrophic as the students are using AI to everything, getting perfect grades while learning nothing.

  9. All of my students expect to get As as I’m at a top university and it costs a lot. They are all smart, but as a result all think they are A+ smart when in reality a lot of them are just B to B+ students at best. They will still go on to medical school though as family wealth paves the way more than ability.

  10. In our local US high-school, all of these numbers apply, but the explanation is not quite what it seems.

    The top 50% of students are still strong. The weak students are weaker than ever. There’s a sharp split along largely demographic lines. To reduce the visible aspects of the “achievement gap” and maintain an acceptable graduation rate, grades outside of honors and AP classes are massively watered down.

    As a result, the awful students pass without knowing how to read, mediocre students get high grades, and good students all get straight A’s.

  11. think these are pretty unrelated tbh, the people with low literacy/numeracy are not in the 40% going to university (or even doing a-levels in the uk, which is also an optional level of education). attainment from among high achievers isn’t ever really going to be affected by the bottom 20%ish

    the only line on the top graph that might be affected by the drop in literacy is the US high school one, which I don’t know enough about to comment on

  12. For UK (the graph below says England,, but if we take it as representative), its actually not so bad? We see improvement on one axis while on the other it stayed the same.

    For the US on the other hand, it is quite telling.

  13. I would like to see this chart next to a similar chart. One axis has the age when kids get their first formal instruction in math and the other for reading.

    And one more that has something about what kinds of homework is assigned, and how often.

    Thise countries in the top righ, Scandinavia and onf more, do things different than the USA dies them. And the USA keeps digging in harder and harder into things not done in the successful countries.

    I have been a teacher in the USA. I have been a student in Norway and Denmark. I have visited schools in Sweden. And I have read abou schools in Finland. It’s amazing how educators ind the USA dismiss the differences.

    It’s like health care. We have lots of evidence. But the people at the top keep claiming the USA is better. It’s almost like someone is getting very rich off the US system. Weird.

  14. The entire graph is a fallacy.

    Correlation is NOT causation.

    There are MANY other factors involved. Even beyond the ones the media likes to occasionally trot out like cell-phones, and the collapse of 2-parent households.

    Just look at the rising poverty levels on the low-end of the student spectrum. The over-regulated and underpaid mess that is the teaching industry, and conflicting interests of State and Local test scores vs actual student skills.

    Education in the US is a giant mess and it’s getting worse not better. That has very little to do with grades.

  15. Mods do your job and get bad math and worse logic out of here.

  16. Holly shit what’s happening in Poland? Am I missing something?

  17. UK school results have gone up by 1 percent in 16 years, including much larger variations within that. It’s not exactly any significant change.

    University results have changed but also a quarter of students in the UK are international students and not everyone enters higher education anyway.

  18. I had no idea the first class degree share in the UK had *doubled* to 30% since I left university. Yale being at 80% A or A- is obscene. 

    What are the specific grade boundaries? 
    I love that a first (70% score) in the UK would be considered a C or something in the US.

  19. Getting smarter and dumber at the same time…

    What a time we live in.

  20. Hmmmm, I wonder which countries might be better tackling this issue —-

    Its who you think it is, it’s like the most boring trope at this point.

  21. The obvious conclusion is that we are all getting smarter and working harder! /s

    Anyways, I wouldn’t put too much weight on the PIAAC, low response rates are a problem.

  22. Oh man this is a truly unpopular opinion:

    But I think good grades are not really an indicator how truly educated you are.
    I’m not in the US, but what I see from outside is not so different from here, even worse it seems (Austria).

    The straight A students in high school were often dumb as bread. They just treated school like a video game. Most efficient way to good grades? Is not to deeply learn the stuff.

    At uni the same thing. I would even make the case, that the people with summa cum laude were the densest, most shallow people overall.
    On the other hand the most curious kids I met often didn’t prioritze school, but focussed on their interests outside of school.

    Later in real life it was the same thing. The ones still proud of their straight As in their 30ies were helpless with real life problems. Almost no critical thinking skills at all.

  23. Only looking at British data. Why doe it say UK on one graph and England on the other? Are these two different studies?

  24. Easter European countries in the bottom left of the graph has to be due to hundreds of thousands of Ukrainian refugees.

    Weird to include this in a graph as some kind of “people are getting dumber” argument. Of course they would have low literacy in a country they just moved into with zero preparation?

  25. Is it the UK or just England? the graph says England which isn’t the same thing

Comments are closed.