The global PISA scores fell almost everywhere this cycle, but not by the same amount, and the gap between the countries that held on and the ones that slid is where the real lesson hides.
When the OECD released the PISA 2025 results this month, the story got flattened into one word: decline. That is true on average, but averages hide the interesting part. A few systems actually gained ground. Several hit their worst scores on record. Reading the split tells you more about what works in education than the global headline ever could.
At a Glance
Most participating countries posted lower reading and math scores in PISA 2025, but the outcomes ranged widely. East Asian systems like Singapore and Japan stayed on top. Türkiye posted the largest gain in the OECD. Germany fell to its lowest since testing began 25 years ago, and Spain and Denmark hit record lows. The countries that held steady tended to share a couple of habits: a return to explicit, structured teaching, and tighter limits on phones in school. For the full picture of the overall decline, see our PISA 2025 overview.
Who Stayed on Top
The top of the table looked familiar. Singapore and Japan, along with other East Asian systems, remained the strongest performers, as they have for years. Within the OECD, Estonia, Finland, Ireland, and Poland held above the average. Stability at the top is its own kind of result in a year when so much slid.
What these systems share is not a secret formula. It is consistency: structured curricula, respected teaching, and a culture that treats sustained academic effort as normal rather than optional.
Who Fell the Hardest
The losers column carried some genuine shocks. Germany recorded its lowest scores since PISA began a quarter-century ago, landing merely at the OECD average in science, reading, and math. Spain posted its worst result since 2000. Denmark hit an all-time low. These are wealthy, well-funded systems, which is exactly why their declines rattled education ministries.
| Country or group | PISA 2025 result |
|---|---|
| Singapore, Japan, East Asian systems | Remained the top performers |
| Estonia, Finland, Ireland, Poland | Held above the OECD average |
| Türkiye | Largest score gain in the OECD |
| Germany | Lowest since testing began 25 years ago |
| Spain, Denmark | Worst national results on record |
| United States | Rank rose as others fell faster; scores near lows (data asterisk) |
The Outlier Worth Studying: Türkiye
Against the tide, Türkiye posted the largest gain of any OECD country. One data point does not prove a model, and this is worth watching rather than copying wholesale, but a system moving up while its peers slide is the kind of case researchers will dig into. When almost everyone is falling, the question flips from “why did scores drop” to “what did the risers refuse to let slip.”
The Uncomfortable US Footnote
The United States looks like a quiet winner if you only read the rankings, since it climbed relative to other countries. Look closer and the picture is less flattering. American teens did not improve much in absolute terms; they mostly held while others fell faster, as Chalkbeat’s analysis explains. US data this cycle also carries an asterisk for incomplete school participation, so treat the specific numbers with care. Climbing a sinking ladder is not the same as climbing.
The Equity Story Underneath the Averages
National averages hide who is being left behind. Germany’s decline, for instance, sits alongside a wide gap tied to students’ socioeconomic background, a pattern that shows up across many falling systems. When scores drop, they rarely drop evenly; the students with the least support at home tend to lose the most. That makes the headline number a poor guide to how healthy a system actually is.
What the Risers Seem to Do Differently
Here is the practical thread, with the honest caveat that this is correlation, not a controlled experiment. The systems that held steady or gained tended to share two moves, noted in European coverage of the results:
- A return to explicit, direct instruction grounded in how learning actually works, rather than chasing the newest classroom trend.
- Tighter limits on phones during the school day, protecting attention where it is hardest to defend.
Neither is a miracle cure, and the deeper drivers of the slide, covered in our look at why PISA scores keep falling, run through every system. But the countries that resisted the trend give ministries something better than despair: a direction.
Frequently Asked Questions
Which country did best in PISA 2025?
Singapore and Japan, along with other East Asian systems, remained the top performers, as in previous cycles. Within the OECD, Estonia, Finland, Ireland, and Poland stayed above average.
Which country improved the most?
Türkiye posted the largest score gain of any OECD country in PISA 2025, standing out as a positive outlier in a cycle where most systems declined.
Why did Germany’s scores fall so much?
Germany recorded its lowest results since PISA testing began 25 years ago, dropping to around the OECD average. Analysts point to a wide achievement gap tied to socioeconomic background, alongside the same attention and instruction pressures affecting other declining systems.
Did the United States really improve?
Mostly in relative terms. US rankings rose because other countries fell faster, not because American scores climbed much in absolute terms. The US data also carries an asterisk for incomplete school participation.
What are the rising countries doing differently?
The systems that held steady tend to share two habits: a return to explicit, structured instruction and tighter limits on phones during the school day. This is a correlation observed across systems, not a proven single cause.
What To Know
The PISA 2025 rankings are less a scoreboard than a set of natural experiments running in parallel. The countries that gained or held did not discover anything exotic; they mostly protected instruction and attention while others let them drift. That is the encouraging read hidden inside a gloomy release: the slide is not a law of nature, and a few systems just proved it.
