Classroom learning outcomes vary sharply by subject, age, geography, and student background. UNESCO estimates that 58% of children reached minimum reading proficiency by the end of primary school, while 44% reached the comparable mathematics threshold in its latest estimate reported in 2026. International assessments also show recent declines in average reading and mathematics performance, alongside substantial differences in instructional time, technology use, attendance, and equity.
Contents
- Global minimum learning proficiency
- Reading achievement and trends
- Mathematics achievement and equity
- Science achievement and trends
- Instructional time and homework
- Digital learning and classroom climate
- Attendance, resources, and teacher conditions
Global minimum learning proficiency
UNESCO’s latest global estimate reported in 2026 separates basic learning outcomes from simply being enrolled or completing a grade. At the end of primary education, 58% of students globally reached minimum proficiency in reading, compared with 44% in mathematics. At the end of lower secondary education, the corresponding shares were 64% for reading and 51% for mathematics. These are global estimates and should not be read as results for every country or classroom.
Completion rates show another part of the picture. In 2024, the global primary completion rate was 88%, compared with 78% for lower secondary education and 61% for upper secondary education. In low-income countries, timely completion was lower: 60% for primary, 37% for lower secondary, and 20% for upper secondary in 2024. Completion therefore does not automatically indicate that students have reached a specified proficiency level.
Regional results can be more severe. Among six sub-Saharan African countries covered by AMPL surveys in 2021 and 2023, about 1 in 10 end-of-primary students reached minimum reading proficiency, while about 2 in 10 reached minimum mathematics proficiency. The AMPL figures apply to those participating countries and survey years, not to every country in the region.
These global and regional measures come from Monitoring education in the SDGs and Monitoring SDG 4. They use the populations and definitions stated by UNESCO.
Reading achievement and trends
PISA provides a different lens: it measures the performance of 15-year-olds in participating education systems. In PISA 2022, Singapore recorded the highest participating-system mean reading score, at 543 points. The OECD average was 476 points. About three out of four students across OECD countries achieved basic reading proficiency, while an OECD average of 7% reached the highest reading proficiency levels, 5 or 6.
The direction of change was generally negative. Across OECD countries, mean PISA reading performance fell by 10 points between 2018 and 2022. In a group of 70 upper-middle- and high-income countries, the share of students proficient in reading fell to 47% in 2022 over the 2012–2022 period. These are assessment trends, not estimates of a single student’s progress.
Change differed considerably between systems. PISA reading scores improved by about 8 to 21 points from 2018 to 2022 in Brunei Darussalam, Panama, Chinese Taipei, Qatar, Japan, the Dominican Republic, and Cambodia. Scores declined by more than 30 points in Albania, Iceland, and North Macedonia over the same period. The comparison covers participating systems with comparable PISA data.
| PISA 2022 reading indicator | Result | Period or population |
|---|---|---|
| Singapore mean score | 543 points | 15-year-olds, 2022 |
| OECD mean score | 476 points | 15-year-olds, 2022 |
| Students at basic proficiency | About 75% | OECD countries, 2022 |
| Students at levels 5 or 6 | 7% | OECD average, 2022 |
| OECD mean change | -10 points | 2018–2022 |
The assessment results are reported in PISA 2022 Results, Volume I, its executive summary, and the OECD analysis of changes in performance and equity.
Mathematics achievement and equity
Singapore also had the highest participating-system PISA 2022 mean mathematics score, at 575 points. The OECD average was 472 points. Across OECD countries, mean mathematics performance fell by almost 15 points between 2018 and 2022, a larger average decline than in reading.
Some systems moved in the opposite direction. Mathematics performance improved by about 10 to 16 points from 2018 to 2022 in Chinese Taipei, Saudi Arabia, the Dominican Republic, Brunei Darussalam, Cambodia, Paraguay, and Guatemala. Scores fell by more than 30 points in Albania, Jordan, Iceland, Norway, and Malaysia.
The distribution of change also matters. Disadvantaged students’ average mathematics performance declined by 17 points across OECD countries between 2018 and 2022, while advantaged students’ average declined by 10 points. The OECD socio-economic gap in mathematics performance widened by 7 points on average during that period. These PISA relationships describe group-level associations and do not establish that one factor caused a particular result.
Science achievement and trends
In PISA 2022, Singapore’s mean science score was 561 points, the highest among participating systems. The OECD average was 485 points. About three out of four students across OECD countries achieved basic science proficiency, and 7% reached science proficiency levels 5 or 6 on average.
Science results were mixed between 2018 and 2022. Performance remained broadly stable in 33 of 71 countries and economies with comparable PISA data, while it improved in 18 PISA countries and economies. Science scores improved by about 15 to 26 points in Kazakhstan, the Dominican Republic, Panama, Chinese Taipei, Japan, Cambodia, and Brunei Darussalam. They declined by more than 20 points in Albania, North Macedonia, Iceland, and Malaysia.
The science results are not interchangeable with the UNESCO minimum-proficiency estimates: PISA describes 15-year-olds in participating systems, while UNESCO reports global or regional populations using its own definitions. The science figures come from PISA 2022 Results, Volume I.
Instructional time and homework
In 2022, students across OECD countries spent an average of 24 hours per week in regular school lessons and 10.8 hours per week doing homework. OECD students also reported an average of 1.5 hours of homework per day during a typical school week.
Homework time was unevenly distributed. Twenty-seven percent of OECD students spent up to 30 minutes per day on homework, 19% spent between 30 minutes and 1 hour, 23% spent between 1 and 2 hours, and 31% spent more than 2 hours per day. These categories describe reported time, not a recommended workload or a causal effect on achievement.
Lesson time also coincided with different mathematics results. OECD students who spent 20 hours or less per week in regular lessons averaged 432 points in PISA mathematics in 2022. This is an observed association across students and systems; it does not show that reducing or increasing lesson time alone produces a specific score.
The time-use findings are from Investments in a solid foundation for learning and well-being: PISA 2022 Results, Volume II.
Digital learning and classroom climate
Digital device use was common but not uniform in OECD schools in 2022. Fourteen percent of students reported no time on digital devices for learning at school, while 31% reported using them for up to 1 hour per day. The reported average was 2.0 hours per day for learning at school and 1.1 hours per day for leisure at school.
The relationship with mathematics varied by amount of use. Students using digital devices for learning at school for up to 1 hour per day scored 25 points higher in PISA mathematics than students reporting no such use. However, the 7.8% of OECD students who reported 5 to 7 hours per day scored 12 points lower than students using devices for 3 to 5 hours. These are associations, not causal estimates, and the groups may differ in other important ways.
Device use also affected the reported classroom climate. Twenty percent of OECD students said they could not work well in most or all mathematics lessons. About 1 in 3 reported being distracted by their own digital-device use in most or every mathematics lesson, and about 1 in 4 reported being distracted by other students’ device use.
The digital-learning and classroom-climate measures come from PISA 2022 Results, Volume II and the OECD overview of learning time and disciplinary climate.
Attendance, resources, and teacher conditions
TIMSS 2023 adds results for younger students. Internationally, 91% of fourth-grade students reached at least the Low International Benchmark in mathematics, compared with 81% of eighth-grade students. In science, the corresponding figures were 90% for fourth grade and 80% for eighth grade. TIMSS percentages describe participating fourth- and eighth-grade students and are not directly comparable with PISA’s 15-year-old population.
Attendance was less consistent in the older grade. In TIMSS 2023, 55% of fourth-grade students and 46% of eighth-grade students reported being absent never or almost never. Just over 10% of students in both fourth and eighth grade reported being absent at least once a week. The attendance figures are self-reported survey results within the TIMSS assessment context.
Teacher preparation is another system-level measure. The share of primary teachers meeting minimum qualification requirements globally was 85% in 2023. In Europe and Northern America, the share was 93%. These figures indicate whether stated minimum requirements were met; they do not measure every dimension of teaching quality or classroom practice.
The TIMSS results are reported in the TIMSS 2023 Press Release, while teacher-qualification measures are reported by UNESCO in Monitoring SDG 4. Across all sources, assessment populations, benchmark definitions, measurement dates, and geographic coverage differ, so the statistics are most useful when their scope is kept explicit.