Performance trends

PISA 2025 marked the ninth round of the international assessment since the Programme was launched in 2000. How do students' results in 2025 compare to those observed in earlier rounds? And what does this mean for the quality of education around the world?

Use the selector above to start exploring trends in students’ performance for a country/economy of interest.

PISA 2025 recorded the lowest OECD-average performance observed so far in science, reading and mathematics. On average across 23 OECD countries, reading has declined after peaking around 2012, while science has declined more gradually over the past decade, and mathematics has fallen especially sharply after 2018.

This chart shows how much a country’s average score has changed compared to changes seen in other countries and economies over a similar period.

Some of the smaller changes are marked in a different colour: these are not statistically significant. In practice, that means one cannot be sure whether they reflect a real change in students’ skills or just the normal variation that can appear over repeated administrations of PISA.

When a change is statistically significant, it is considered large if it is greater than 20 score points. To give a sense of scale: 20 points is roughly the amount of learning that 15-year-olds typically gain in one school year across OECD countries. So, if a country’s score falls by 20 points between two PISA assessments, it is as if students were performing at the level expected of 14-year-olds in the previous cycle.

It is important to remember that changes seen in PISA may have their roots in earlier years – before PISA started measuring them. For example, education reforms often take time to show effects, and investments in early childhood education may only become visible a decade later when those children reach 15. 

On average across OECD countries, the proportion of students performing at the highest level of science proficiency measured by PISA did not change significantly between 2015 and 2025, while the proportion of students performing at or below the lowest level of proficiency increased.

Changes observed within a country/economy may not have affected everyone equally. Select a subject to explore how the average performance of boys and girls changed over the most recent decade. Gender differences that are shown in white are not statistically significant.

Socio-economic disparities within a country are often larger than the differences in performance that are observed between countries/economies. Select a subject to explore how the gap in mean performance between the 25% most advantaged and the 25% most disadvantaged students changed over the most recent decade.