PISA 2025: Maths, Reading and What the Data Shows

I have a friend who lectures at university, and he is a fascinating man. When he lived in Vietnam, he saw Western education as the light at the end of the tunnel: modern, advanced, civilised and deeply practical. Bring it over wholesale, he thought, and Vietnam would surely rise like a tiger. Then he settled and began teaching in Australia, and his view flipped completely: "The kids here can't do maths, they can't read a question properly, and all they're good at is… smiling it off."

To be fair, my friend is not alone. Many Vietnamese people, once they come into contact with education abroad, fall into one of two extremes: they either idolise it or dismiss it entirely. Both are equally risky, because both come from personal feeling rather than objective data.

PISA does not measure how hard the material you have studied is. It measures whether you can use maths to deal with a real situation.

Look at the PISA 2025 results, the OECD's global assessment of 15-year-olds published on 8 September 2026, and a different picture emerges. In maths, China's four participating provinces and municipalities (Beijing, Shanghai, Jiangsu and Zhejiang) lead with 612 points, followed by Singapore on 563, Macao 549, Chinese Taipei 546 and Japan 525. In Europe, Estonia scores 508 and Switzerland 499. Meanwhile, the "English-speaking countries that are bad at maths", the UK (488), Canada (485), New Zealand (480) and Australia (478), are still ahead of Vietnam (443). Even the United States, on 463 points, exactly the OECD average, ranks above us.

Nor is this a one-off. In PISA 2022, Canada, the UK, Australia and New Zealand were all ahead of Vietnam too.

Before going further, one important point: Vietnam's 2025 and 2022 scores cannot be compared. In 2022, Vietnamese students sat the test on paper. 2025 was their first time taking it on a computer. When the test format changes, the scores are no longer on the same ruler, which is why the OECD itself does not use these two numbers to say whether Vietnam has improved or declined. So please don't look at 469 falling to 443 and conclude that Vietnam has "slipped down the rankings".

So where does the feeling that "Westerners aren't good at maths" come from?

The answer is that school maths in English-speaking countries (the US, the UK, Australia, Canada…) really is lighter than ours, so much so that a strong Year 9 pupil in Vietnam can sometimes solve most of the SAT maths paper (the standardised test used for US university applications).

That is because these countries do not make advanced maths compulsory. In England, for example, maths is compulsory until 16, while advanced maths at A level is each student's own choice. School maths there is designed for the many and puts application first. Not every pupil is made to prove three-dimensional geometry, wrestle with complex calculus or "swim" through higher algebra. Students in Japan, Korea, Singapore, China and Vietnam, by contrast, usually meet much heavier maths early in secondary school.

But here is the point worth pausing on: a lighter curriculum does not mean weaker students. PISA does not measure how hard the material you have studied is. It measures whether you can use maths to deal with a real situation: reading a chart, comparing two phone plans, estimating a risk. And on that measure, students from the "light maths" countries are still ahead of us.

The data points to something even more thought-provoking. If studying heavier maths made us better at it, we should at least have more top performers. The reverse is true: only about 4% of Vietnamese students reach the top level in maths (Levels 5–6), half the OECD average of 8%. At the other end, 58% reach baseline proficiency, below the OECD's 65%.

What we lack may not be maths at all

Remember my friend complaining that Australian students "can't read a question properly"? Here is what PISA 2025 says: Australia scored 491 in reading. Vietnam scored 392.

A numeracy lesson at HEI Schools Saigon Central

That is the number that really deserves our attention. Our reading score is 69 points below the OECD average, the widest gap across all four areas assessed. Only 41% of Vietnamese students reach baseline proficiency in reading, compared with 58% in maths. Our maths score is about 50 points higher than our reading score, one of the most lopsided profiles anywhere, alongside Macao and Azerbaijan.

Why does this matter for maths itself? Because a PISA maths question is almost never "solve this equation". It is a passage of text: a shopping situation, a chart, a price list, and the question "which option should you choose, and why?" A student may be excellent at calculus, but if they cannot work out what the question is asking, that knowledge is left standing outside the door. We sharpen the knife with great care, yet many children still cannot read the recipe.

In fairness, one more thing needs saying: 2025 was Vietnam's first computer-based test, and the reading section required students to read digital texts, move between several screens and interpret dynamic tables and charts. Part of the gap may come from on-screen reading skills, rather than simply "children not reading books". But whatever the cause, the message stays the same: reading is the foundation of every subject, maths included.

… but PISA is only the surface

Mathematics in the West truly comes into its own at university.

At that level, the educational philosophy is turned completely on its head: "Mathematics is not for everyone." Only genuinely passionate and capable students can keep up. And the irony is that the very best mathematicians in the world, from Asia and continental Europe, travel to Western universities to study, research and invent.

Nobel and Fields prizes, and most breakthrough scientific theories, still largely belong to the United States and Western Europe. That is because, however light their schooling may be, their universities and research ecosystems are extremely demanding. We, in my view, sometimes push academic maths all the way down into primary school, leaving many pupils without direction, without motivation, and in the end without belief in themselves.

So saying "Western school maths is lighter" is right if we are comparing how difficult the school curriculum is. But saying "Westerners aren't good at maths" is wrong, even on average, and even more wrong when we look at the summit. My friend once complained that the West is "bad" at maths, which is why it has to "import" talent from developing countries. In fact, it is a wise strategy. They know how to identify and make use of the world's best minds. The proof is that the talent they "attract" tends to blossom precisely within the education ecosystem they have built.

Of course, the West has its problems too…

The gap in investment is vast. In 2024, Vietnam's state budget for education and training was about 380 trillion dong (roughly US$15 billion), for more than 23 million preschool and school pupils, not counting university students. Australia, with only about a quarter of Vietnam's population, spent more than A$110 billion on education at all levels (public and private combined) as far back as 2015. The two figures are not calculated in the same way, but they are enough to show the scale of the difference. And yet the return is not necessarily as far apart: Australia's PISA maths score has fallen 46 points since 2003 to its lowest level ever, and the country is also grappling with a shortage of teachers.

Looking more widely, PISA 2025 recorded the lowest OECD average ever in all three subjects. In maths alone, the OECD average fell by 22 points between 2015 and 2025, equivalent to more than a year of learning.

And that trend reaches even the place that many Vietnamese, my friend included back when he lived in Vietnam, once held up as the model: Finland. Three years ago, Finnish students scored 484 in maths; this year, 469, only a little above the OECD average. Compared with 2015, for every 100 students there are now 17 more who fall short of baseline proficiency in maths.

Global PISA scores in maths, reading and science have been steadily declining over the past 20 years

Last year, I wrote that Finland's problem lay in the gap between good schools and weak ones. This year's data makes me correct that: the differences between Finnish schools remain fairly small; the gap that is widening sits inside each school, between students who are doing well and those who are struggling. I work with a Finnish curriculum, so writing these numbers down is far from comfortable. But that is exactly why I have to write them. Idolising a model, believing that doing everything exactly as Finland does will make our children excel, is also a way of letting feelings lead, just like my friend back in his Vietnam days.

No model is immune to difficulty, and no model should be transplanted wholesale from one place to another. The job of an educator is to keep the core that science supports, such as learning through play, respecting each child's own pace of development, and trusting and empowering teachers, and then to adapt the rest to Vietnamese children and families. And just as important: having the courage to look the data in the eye, even when it is not as pretty as we had hoped.

The question is not whether the West is better or we are cleverer, but which lens we are looking through

If we learn from others, we should learn selectively. If we criticise, we should criticise with evidence. And if we use only our feelings as the measure, we stay stuck in an intellectual game whose losing side is… our own understanding.

Vietnam has a precious asset: a deep love of learning. But for that love of learning to become national strength, what we also need is a practical philosophy of education, one that respects time, streams sensibly and, above all, does not insist that everyone must excel at maths. Maths is a sophisticated weapon. Hand it to everyone without a strategy, and we may only raise a generation that wields a butcher's cleaver… to peel an orange: exhausting, off-target, and without understanding what they are doing.

And what matters most is not on any league table, but whether, after fifteen years of schooling, a child still wants to read, to ask and to learn.

And if I had to choose one place to start, I would not choose another extra maths class. I would choose a book. The OECD's analysis of PISA data has shown that 15-year-olds whose parents read to them from their early years into primary school scored markedly higher in reading, even when comparing families of similar socio-economic background.

Many studies following children from preschool also show that early language skills predict later achievement in maths. That is a correlation, not a certainty. But in the preschool classrooms at HEI Schools Saigon Central, maths still always begins with a story: a rabbit counting carrots, a bear sharing a cake fairly.

Because before numbers live in equations, they live in stories. And what matters most is not on any league table, but whether, after fifteen years of schooling, a child still wants to read, to ask and to learn.