OECD Finds Reading and Math Scores at Record Lows
The OECD reported on September 8, 2026, that average reading and mathematics performance had fallen to the lowest levels recorded by the Programme for International Student Assessment, or PISA, across OECD education systems.
The results cover more than 760,000 15-year-old students in 91 countries and economies, representing about 33 million young people. They show a broad decline in foundational skills while also highlighting differences in socioeconomic conditions, family engagement and how students use digital tools.
What changed in the 2025 results
Across OECD countries, average reading scores fell by 28 points between 2015 and 2025. The OECD describes that decline as roughly equivalent to one and a half years of learning on its scale. Mathematics scores fell by 22 points, or just over one year of learning.
The decline was smaller among non-OECD participants over the same period: 16 points in reading and eight points in mathematics. That distinction matters because PISA is not a ranking of every school system in the world. It compares participating countries and economies, with OECD averages serving as one specific comparison group.
PISA measures how 15-year-olds apply knowledge and skills to problems, rather than testing only whether students have memorized school content. The results therefore provide an international view of performance in reading, reasoning and quantitative understanding, but they are not national school-exam results.
One in five low performers across three subjects
Across OECD countries, 20% of 15-year-olds were estimated to be low performers in science, mathematics and reading, up from 16% in 2022. The figure does not mean that one in five students failed a national school exam. It refers to students below PISA’s baseline proficiency thresholds in all three domains.
The OECD says students at that level lack foundational skills needed for further education, employment and everyday life. For schools, the finding raises practical questions about early intervention, focused curricula, teacher support and targeted help for students who have fallen furthest behind.
Inequality remains central
Socioeconomic background continued to be strongly associated with performance. Across OECD countries, advantaged students scored 85 points higher than disadvantaged students in science.
The gap narrowed between 2022 and 2025, but the OECD said that change was driven largely by weaker performance among advantaged students rather than gains among disadvantaged students. A smaller gap, in other words, does not necessarily mean that students most in need improved.
Family engagement also appeared to be associated with stronger results. Students who said a family member asked at least weekly what they had done in school scored about 35 science points higher than students reporting less family interest, according to Associated Press reporting on the OECD findings. The comparison included adjustments for socioeconomic differences, but it does not prove that one family behavior causes a particular score.
Digital distraction and AI show mixed results
More than one in four students said classmates were distracted by digital devices in most or every science lesson. Students surrounded by that distraction performed less well, reported lower engagement and felt less connected to school. The OECD also linked excessive device use and leisure-oriented use at school with weaker performance.
Those findings do not establish that devices caused the international decline. They point instead to the importance of purpose, moderation and classroom conditions.
The same caution applies to artificial intelligence. Students who used AI to help them learn and were taught to assess the quality of AI-generated information tended to perform better than students who received no such guidance. Students who said they did not use AI to draft writing assignments outperformed those who reported that practice.
These comparisons are based on reported use patterns, not a controlled experiment. They do not show that AI is inherently beneficial or harmful. They suggest that schools will need to distinguish between tools that support students’ thinking and tools that replace it.
Decline is broad, but not universal
The OECD said some systems improved in science between 2015 and 2025, including Costa Rica, the Slovak Republic, Türkiye and the United Kingdom among OECD participants, along with several non-OECD participants. Most participating Chinese jurisdictions, Estonia, Japan, Korea, Singapore, Chinese Taipei and the United Kingdom were among the highest-performing systems across science, mathematics and reading.
Those differences indicate that the overall downward trend is not inevitable, even though the OECD averages show a serious international problem. The results also caution against treating one global explanation as sufficient: the weakening of performance began before the COVID-19 pandemic in many countries, and the report describes longer-term relationships involving digitalization, screen time, reading for enjoyment and socioeconomic conditions.
What happens next
The practical policy questions are familiar but urgent: how to strengthen foundational reading and mathematics, support teachers, engage families, target assistance and keep technology focused on learning.
The OECD reported the evidence and identified possible areas for action; it did not order governments to adopt a single response. For parents and educators, the clearest message is that technology and AI are not presented as automatically good or bad. Their value depends on how they are used, whether students remain engaged in the thinking, and whether schools protect time for focused learning.
Sources
- OECD: PISA 2025 performance release
- Associated Press: Student scores in high-income countries hit a low point
- Deutsche Welle: PISA 2025 results
Look for updates to this story
Discover more from Interactive News
Subscribe to get the latest posts sent to your email.