What is PISA, exactly?

PISA — the Programme for International Student Assessment — is a survey run by the OECD every few years that tests a representative sample of 15-year-olds in reading, maths and science, and compares average performance across participating countries and regions.

Does PISA test every child in a country?

No. PISA tests a statistically designed sample of schools and students within each participating country or region — a few thousand pupils, not the whole cohort — and uses that sample to estimate a national average. It is a survey, not a census or an exam every student sits.

Does England, Scotland, Wales and Northern Ireland get one PISA score or four?

Each UK nation is sampled and reported on separately in PISA, because each runs its own education system and curriculum. There is no single combined 'UK exam board' score, and comparing one nation's PISA result to another's needs the same care as comparing their exam systems generally.

Does a country's exam board structure explain its PISA rank?

Not in any way that has been convincingly established. Countries with very different assessment structures — from Finland's minimal standardised testing to Singapore's centralised national exams — have both scored highly on PISA at different points, which is itself evidence against any simple structural explanation.

Is TIMSS the same thing as PISA?

No, and confusing the two is a common source of error. TIMSS is run by the IEA, not the OECD, tests pupils in Grade 4 and Grade 8 rather than 15-year-olds, and assesses mastery of the mathematics and science curriculum actually taught, rather than PISA's applied, curriculum-independent literacy skills. A country's TIMSS and PISA rankings can genuinely diverge because the two studies measure different things in different pupils.

What PISA rankings can and can't tell you about exam systems

PISA rankings are a useful, but frequently over-read, snapshot: the OECD tests a sample of 15-year-olds in reading, maths and science every few years and ranks participating countries by average score. PISA is not an exam any individual student sits, it has no relationship to any exam board, and it cannot, by itself, tell you whether one country's exam-board structure is "better" than another's.

Headlines about PISA rankings tend to arrive with strong implied conclusions — that a high-ranking country's whole education system, including how it examines pupils, must be doing something the rest of the world should copy. The OECD's own documentation is considerably more cautious than the headlines, and the gap between what PISA measures and what gets claimed about it is worth understanding before drawing any conclusion about exam boards specifically.

What PISA actually is

The Programme for International Student Assessment (PISA) is run by the OECD (Organisation for Economic Co-operation and Development) roughly every three years — though the cycle has occasionally shifted, including a delay caused by the COVID-19 pandemic. It tests 15-year-olds in participating countries and economies on reading, mathematics and science, rotating which subject gets the most detailed focus in each cycle.

Two features of its design matter enormously for how the results should be read:

It is a sample, not a census. PISA does not test every 15-year-old in a country. It uses a statistically designed sample of schools and, within them, a sample of students, chosen to be representative of the national population at that age. National averages are estimates derived from that sample, with associated margins of error — not a direct measurement of every child.
It tests skills, not any national curriculum or qualification. PISA questions are designed by the OECD to assess general reading, mathematical and scientific literacy — the ability to apply knowledge to unfamiliar, real-world-style problems — rather than testing mastery of any specific country's curriculum content. A country's GCSE, Abitur, Baccalauréat or any other national qualification plays no role in PISA's design or marking.

Why this rules out a simple exam-board explanation

Because PISA is not linked to any national exam or qualification, it cannot be a direct measurement of how well an exam board does its job. A country's PISA average reflects, at most, general literacy and numeracy skills in its 15-year-old population at one point in time — filtered through sampling, translation, item design and participation rates that vary between cycles. It says nothing about how many exam boards that country has, whether they are charities or companies, how papers are marked, or how grade boundaries are set.

This distinction matters directly for how the UK looks in PISA data. England, Scotland, Wales and Northern Ireland are reported as separate entities within PISA, exactly because each has its own curriculum and its own assessment system — see our companion articles on why Scotland has the SQA, why Northern Ireland kept CCEA and why Wales created Eduqas. There is no single "UK exam board score" in PISA data to compare against any other country, only four separate national results sitting inside the same set of islands.

The structural diversity among high-PISA-scoring countries

If exam-board structure drove PISA performance in any straightforward way, you would expect high-scoring countries to share a similar structure. They don't. Countries that have performed strongly on PISA over various cycles include:

Highly centralised systems with a single national exam board or ministry-run assessment, such as Singapore and South Korea.
Systems that historically minimise external standardised testing at younger ages and rely far more on teacher-based, continuous assessment until a single matriculation exam near the end of upper-secondary school, such as Finland.
Federal or regional systems with multiple boards or regional variation in assessment, in various OECD countries.

This spread — from minimal external testing to heavily centralised national exams, both represented among strong PISA performers at different times — is itself evidence against treating "number of exam boards" or "how centralised assessment is" as a clear driver of PISA results. If both ends of that spectrum can produce strong performance, the explanation for any individual country's result is very unlikely to be its exam-board structure in isolation.

What actually might explain differences (with appropriate caution)

Researchers who study PISA results point to a wide range of possible contributing factors — teacher training and status, funding and how equitably it is distributed, poverty and inequality, the amount and nature of early-years education, cultural attitudes to schooling, and the amount of out-of-school tutoring, among others. None of these has been established as a single, dominant cause, and PISA's own documentation is explicit that its results should not be read as ranking entire education systems on a single dimension of quality. Anyone citing a specific causal claim about why a particular country scores well should be able to point to a specific study, not just the ranking itself.

PISA is not the only international assessment — and confusing it with TIMSS causes real errors

A specific, common mistake is treating any international ranking of school performance as if it were PISA. The other major study parents are likely to see cited is TIMSS (Trends in International Mathematics and Science Study), run by the IEA (International Association for the Evaluation of Educational Achievement) rather than the OECD, and it differs from PISA in ways that matter for how its results should be read too:

TIMSS tests curriculum content, not general literacy. Where PISA deliberately avoids testing any single country's curriculum and instead assesses applied reading, maths and science literacy, TIMSS is explicitly designed around the mathematics and science curricula that participating countries actually teach — it is closer to "how well did pupils learn what they were taught" than PISA's "how well can pupils apply general skills to unfamiliar problems."
TIMSS tests different ages. It assesses pupils in Grade 4 and Grade 8 (roughly ages 9-10 and 13-14), rather than PISA's fixed cohort of 15-year-olds — so a country's TIMSS and PISA results are describing different pupils at different stages of school, not the same students measured twice.
The two rankings can, and do, diverge. A country strong on TIMSS (curriculum mastery at 9-10 or 13-14) is not guaranteed to rank the same way on PISA (applied literacy at 15), precisely because the two studies are measuring genuinely different things, run by different organisations, on different age groups, using different item designs.

A headline that cites "international rankings" without naming which study it means, or that treats a TIMSS result as if it were a PISA result (or vice versa), should be read with real caution — the two are not interchangeable, and conflating them is one of the most common ways PISA-adjacent claims go wrong.

Participation is voluntary, and that shapes the results too

PISA is not compulsory for a country to join, and which countries and economies choose to participate — and how consistently they do so cycle after cycle — is itself part of the context needed to read any single ranking. Recent PISA cycles have involved participation from roughly 80 countries and economies, a mix of OECD member states and non-member partner countries and regions, and that list has changed somewhat between cycles as countries join, withdraw, or have participation affected by events (the 2022 cycle's own timing, for instance, was shifted by the COVID-19 pandemic, as already noted above). A country not appearing in a given cycle's table isn't necessarily performing badly — it may simply not have participated in that round — and a change in a country's rank between cycles can reflect a change in which other countries took part, not only a change in that country's own pupils' performance.

What PISA is genuinely good for

None of this means PISA is not useful — it is one of the few large-scale, methodologically consistent international comparisons of student skills available, run to a published, peer-reviewed methodology, and it gives policymakers a way to track how a country's average 15-year-old reading, maths and science literacy is trending over time relative to comparable countries. That is a genuinely valuable, and genuinely limited, thing to measure.

The limitation is precisely the point of this article: PISA measures literacy skills in a sampled population at a single age, using OECD-designed items unconnected to any national qualification. It is not, and was never designed to be, a report card on any country's exam boards, grading systems, or the number of awarding bodies it happens to run. Anyone comparing UK exam boards to those of another country, on their own structural merits, needs different evidence than a PISA table — and should be wary of anyone offering a PISA rank as if it settled the question.

The precise conclusion

A high PISA rank tells you a country's sampled 15-year-olds performed well, on average, on a specific set of internationally designed literacy items, in a specific testing cycle. It does not tell you that country's exam boards are better run, its grading is more reliable, its qualifications are more rigorous, or that any particular feature of how it organises assessment caused the result. Treat PISA rankings as what the OECD designed them to be — a skills snapshot — and look elsewhere, to the structural and regulatory detail covered across this cluster, for anything resembling an answer about exam-board quality itself.