International student assessments help governments identify where students are succeeding, where they are falling behind and how learning outcomes are changing over time.
The Programme for International Student Assessment, or PISA, tests how well 15-year-olds can apply what they have learned in science, mathematics and reading to real-world problems. Run by the Organisation for Economic Co-operation and Development (OECD), the assessment is intended to identify weaknesses and track whether learning improves over time.
The results released on September 8 show improvement in some areas, declines in others and significant gaps in the regional data. PISA 2025 placed particular emphasis on science and introduced a new assessment of computational problem-solving, part of its broader work on learning in the digital world.
What the Results Show
Kazakhstan, Kyrgyzstan, and Uzbekistan participated nationally. Tajikistan was represented by Dushanbe rather than a national sample, while Turkmenistan did not participate.
| Participant | Science | Math | Reading | Comp. Problem Solving |
| Kazakhstan | 420 | 414 | 385 | 449 |
| Kyrgyzstan | 363 | 364 | 344 | 388 |
| Dushanbe, TJK | 334 | 382 | 368 | — |
| Uzbekistan | 438 | N/A | N/A | 402 |
| OECD average | 482 | 463 | 461 | 500 |
Kazakhstan recorded the highest published regional scores in mathematics, reading and computational problem-solving, while Uzbekistan recorded the highest score in science. Uzbekistan’s mathematics and reading results were not published, preventing a complete comparison across the three core subjects.
Central Asia remained below OECD averages, but Kazakhstan and Uzbekistan scored higher than several EU member states in some subjects.
Across OECD countries, average mathematics and reading performance fell to the lowest levels yet recorded by PISA. Science was comparatively stable.
What Changed
The latest scores and the longer-term trends do not always point in the same direction. Kazakhstan and Uzbekistan both participated in PISA in 2022, while Kyrgyzstan’s previous participation was in 2009.
| Country | Comparison Period |
Science | Math | Reading |
| Kazakhstan | 2022–2025 | −4 | −12 | −1 |
| Kyrgyzstan | 2009–2025 | +33 | +33 | +30 |
| Uzbekistan | 2022–2025 | +83 | N/A | N/A |
Kazakhstan’s mathematics score fell by about 12 points from 2022, a statistically significant decline for the country. Its smaller declines in science and reading were not statistically significant.
Kyrgyzstan improved in mathematics, reading and science compared with its previous participation. The gains cover 16 years rather than a single PISA cycle.
Uzbekistan recorded the largest movement in the regional results. Its science score rose from 355 in 2022 to 438 in 2025, an increase of 83 points. That was also the largest science gain among education systems with comparable results worldwide and more than twice the next-largest increase of 38 points.
What Other Assessments Show
Other international assessments help put the larger changes in context.
Kazakhstan and Uzbekistan were the only Central Asian countries to participate in the 2023 Trends in International Mathematics and Science Study, or TIMSS. The assessment tests younger students and uses a different methodology from PISA.
Among eighth-graders, Kazakhstan scored 443 in science and Uzbekistan 396. Two years later, PISA recorded science scores of 420 for Kazakhstan and 438 for Uzbekistan. The assessments cover different students and testing methods, so the scores cannot be compared directly.
Comparable evidence is thinner elsewhere in the region, making it harder to cross-check changes consistently across all five countries.
Gaps in the Data
Uzbekistan’s PISA 2025 mathematics and reading scores were not published. The OECD said it found inconsistencies in responses in some schools and regions, making those results unsuitable for comparison. Its science and digital-learning results were published.
Kyrgyzstan’s 16-year gap between assessments makes recent trends harder to establish. Dushanbe’s results cannot be treated as a national measure of Tajikistan, while Turkmenistan’s absence leaves no PISA benchmark.
Central Asia therefore still lacks a fully comparable set of results covering all five countries.
Measuring Improvement
Regular and credible testing gives governments evidence to identify weaknesses, track outcomes following reforms, and decide where additional resources may be needed.
PISA 2025 provides a new benchmark for Central Asia. The next round will show whether the patterns seen in 2025 will continue.
