Study

PISA 2025: score declines and the limits of interpretation

Lower average scores do not, by themselves, tell us why achievement has declined. This study examines comparisons across the 2018, 2022 and 2025 PISA rounds and the evidence used to explain their differences.

Data and approach

The analyses draw on common test items and response times, measurement models, socioeconomic and population-composition comparisons, TIMSS pupil panels, and other international and national assessments. They examine whether proposed explanations fit the patterns across countries, subjects and groups of pupils.

Ten explanations examined

Social media use
Screen time and school phone restrictions
AI use and the socioeconomic pattern
Curriculum reform
Pandemic disruption and school closures
Migration and changing pupil composition
Rapid responding and test effort
Changes in assessment methods
Attention and reading comprehension
Disengagement and pupil profiles

What the evidence can establish

The comparisons provide factual counterarguments to simple, single-factor explanations. They do not establish that any proposed factor is irrelevant or estimate how much each caused the decline. Country-level associations, descriptive adjustments and intervention studies answer different questions.

Substantial reading and mathematics declines remain in several analytical specifications, with differences across countries and groups of pupils.

Published analytical brief

Interpreting the PISA score declines: evidence and cautions for ten proposed explanations

Download the analytical brief (PDF)

Zenodo: supporting data and charts

Milos Kankaras, 2026. The brief contains 15 pages and ten charts. It is an independent analytical report, not a peer-reviewed article or an OECD publication.