# PISA 2025: every country's report card

*English edition: https://mariuscomper.uk/lumea-digitala/en/ · Ediția în limba română: https://mariuscomper.uk/lumea-digitala/*

An interactive dashboard, for any of the 91 countries and economies in PISA 2025, across eight themes: performance, socio-economic equity, gender gaps, AI and digital learning, engagement, school climate, family support and the environment. Single source: OECD (2026), *PISA 2025 Results (Volume I): Future-Ready Students*, https://doi.org/10.1787/73451bc5-en.

## The opening figure: Romania

The science-score gap between advantaged and disadvantaged students is 122 points in Romania, against an OECD average of 85 — the fourth-largest gap of 90 countries with data (Table I.2). Even so, 72.6% of Romanian students say they are taught in lessons to check the quality of AI-generated information — above the OECD average of 62.6% (Table I.6) — and Romania has one of the OECD's strongest family-support climates: second-highest, after Albania (Table I.7).

PISA test outcomes remain well below average all the same: 425 points in science (OECD average: 482), a statistically significant 16-point drop in reading since 2022, and 38.5% of students below the basic proficiency level (Level 2) in science, reading or mathematics — nearly double the OECD average of 19.7% (Table I.1). PISA does not establish a causal link between these facts; all are shown, sourced, so the reader can see the tension.

## How each country dashboard reads

Every country in the interactive tool has eight sections, each compared against the OECD average:

1. **Performance** — mean scores in science, reading, mathematics and computational problem solving, the change since PISA 2022 (flagging statistically significant changes), and the shares of top performers and below-basic-level students.
2. **Socio-economic equity** — how much a student's score depends on their socio-economic background, the score gap between advantaged and disadvantaged students, and the share of disadvantaged students who succeed anyway.
3. **Gender gaps** — the boys-minus-girls score gap in science, reading and mathematics.
4. **AI and digital learning** — school capacity to use digital devices, share of students using AI at least weekly to learn, share taught to assess AI-generated information, freedom from distraction, phone bans and subject-specific guidelines.
5. **Engagement and attitudes** — perseverance, curiosity, goal setting, help seeking, beliefs about science, cognitive adaptability, growth mindset and hasty reading responses.
6. **Climate and safety** — disciplinary climate, absence of bullying, absence of truancy, sense of belonging, perceived value of school and belief in social mobility.
7. **Family and school support** — family and teacher support, absence of teacher shortages, peer-to-peer tutoring, absence of material shortages.
8. **The environment** — environmental science score, the socio-economic gap on that test, students' belief they can generate change, participation in environmental activities, and interest in an environment-related career.

Six countries/economies (Canada, the Netherlands, New Zealand, Norway, Albania, the United States) carry an interpretation caution, explained in the tool and in the table below.

## A sample of the 91 dashboards

| Country | Science | Socio-economic gap (points) | Boys − girls gap in reading | Use AI weekly to learn |
| --- | ---: | ---: | ---: | ---: |
| OECD average | 482 | 85 | −30 | 45.5% |
| Romania | 425 | 122 | −30 | 39.9% |
| Japan | 538 | 71 | −14 | 27.4% |
| Singapore | 560 | 91 | −26 | 65.6% |
| United Arab Emirates | 458 | 50 | −37 | 64.3% |
| United States* | 502 | 104 | −24 | 37.3% |

*The United States has limited reporting for questionnaire-based figures (see above).*

The full tool, with all 91 countries, a chart on the metric you pick, and a direct compare between any two, is at https://mariuscomper.uk/lumea-digitala/en/.

Data licence: CC BY 4.0. Full sources and method notes: see the main page's "Method and sources" section.