PISA 2025 Low Performers: One in Five, and What Helps

The number of PISA 2025 low performers is the finding that should worry parents more than any ranking. Across OECD countries, 20% of 15-year-olds now score below Level 2 in science, reading, and mathematics simultaneously, up from 16% in 2022. The OECD’s description is blunt: “These students lack the foundational skills needed to succeed in further education, employment and everyday life.”

One in five. Not one in five who struggles with algebra. One in five who cannot reliably read a moderately complex text, interpret a simple data table, or use basic scientific reasoning.

The interesting part is that five education systems moved the other way over the past decade, shrinking their low performers while growing their top performers. What they have in common is not money.

Key Takeaways

• 20% of OECD 15-year-olds are low performers in all three subjects, up from 16% in 2022. “Low performer” means below Level 2, PISA’s baseline proficiency.
• Over the decade, low performers rose about 10 percentage points in reading and 9 in math across OECD countries, while top performers fell about 2 points in each.
• Only Montenegro, Qatar, the Slovak Republic, Türkiye, and the United Arab Emirates cut low performers and raised top performers over the past decade.
• Spending is a weak predictor. Luxembourg spends USD 288,521 per student and scores below Ireland, France, and Italy in science, which each spend under half that.
• Socio-economic background accounts for 12% of within-country variation in science performance, which means it matters and is far from destiny.

What “PISA 2025 low performers” actually means

In the United States, NCES administers PISA and reports the same proficiency levels. A low performer is a student scoring below Level 2 on a subject scale, which the OECD treats as baseline proficiency. In science, Level 2 means a student “can use scientific knowledge and evidence to explain familiar phenomena and interpret straightforward data.” About three-quarters of students across the OECD reach at least that bar in science. The headline 20% figure is stricter: it counts students below Level 2 in all three core subjects at once.

Volume I, released September 8, 2026, tracks both tails. Over the past decade across OECD countries:

• Science: low performers up about 3 percentage points; top performers (Level 5 or 6) unchanged.
• Reading: low performers up about 10 percentage points; top performers down about 2.
• Mathematics: low performers up about 9 percentage points; top performers down about 2.

The distribution is stretching downward. That matters more than a sliding average, because the consequences of falling below Level 2 are not linear. A student at Level 3 who slips to Level 2 has a worse year. A student who slips below Level 2 has a harder adulthood.

The spread within countries varies enormously. In science, the gap between the 90th and 10th percentile averages 258 points across the OECD, but exceeds 290 in Luxembourg and the United States. We looked at that US pattern in why the US rank rose while scores stayed flat.

Common beliefs about struggling students versus what PISA 2025 shows

Parent conversations about struggling kids run on a handful of assumptions. Here is how each holds up against this data set. I have been deliberate about labeling where the evidence is thin.

Common belief	What PISA 2025 data show	Strength of evidence
”More money per student fixes it”	National wealth accounts for about half the variation between countries; Luxembourg spends USD 288,521 per student and trails Ireland, France, and Italy in science	Strong
”It’s mostly the pandemic”	Declines predate COVID in many countries; closure length does not predict reading loss (Sweden: no closures, −21 points)	Strong
”Poor kids can’t reach proficiency”	Socio-economic status explains 12% of within-country variation in science; B-S-J-Z (China), Canada, Estonia, Ireland, Japan, Korea, and Macao combine fairness with high performance	Strong
”Discipline is a side issue”	Disciplinary climate improved in 37 systems between 2015 and 2025, and that is “associated with stable or improved science performance”	Moderate (correlational)
“Struggling kids just need more effort”	Perseverance correlates with performance in nearly every system, but the OECD does not establish direction of causation	Moderate
”Growth mindset changes outcomes”	Students reporting a growth mindset score 28 points higher in science after controls, with the biggest gaps among mid-performers	Moderate (association)
“It’s a phone problem”	28% report classmates distracted in most science lessons and phone policies correlate with less distraction, but no causal estimate on scores	Thin

That last row deserves emphasis. Phones are the most popular single explanation in parent circles, and the honest answer is that PISA cannot isolate their effect on scores. It can show distraction is widespread and that clear school policies coincide with less of it.

What the five improving systems appear to share

Montenegro, Qatar, the Slovak Republic, Türkiye, and the United Arab Emirates were the only systems that reduced low performers while increasing top performers over the decade. Twelve others reduced low performers without the top-end gain.

The OECD does not claim a causal recipe, and I will not invent one. But its preface names the levers it considers most promising, and they are strikingly unglamorous:

Focus. “Set high, consistent expectations for every student in reading, mathematics and science.” Also: “favour depth over breadth, better to teach a smaller set of core topics well enough for genuine mastery than to rush through a long list only superficially.”

Alignment. “Curriculum, teaching and assessment need to reinforce each other rather than pull in different directions.”

Order. Classrooms “kept free of avoidable digital distractions.” TALIS 2024 found teachers spending 16% of class time keeping order, up from 13% in 2018. That is roughly seven minutes of a 45-minute lesson.

Teaching skill, not just subject knowledge. Teachers need “the practical skills to explain clearly, manage a classroom and keep students engaged and motivated.”

Resources matched to need. Targeted at “disadvantaged students and schools” rather than spread evenly.

One more finding that is unusually actionable: “The provision of study help, such as peer-to-peer tutoring, is associated with higher science performance.” That aligns with the strongest evidence base in the whole literature on struggling students, which the Institute of Education Sciences summarizes for schools. Nickow, Oreopoulos, and Quan’s meta-analysis of 96 randomized trials (NBER Working Paper 27476, 2020) found a pooled effect size of 0.37 standard deviations for tutoring, with teacher and paraprofessional tutors outperforming nonprofessional and parent tutors, and reading gains concentrated in earlier grades while math gains showed up later.

The uncomfortable finding about narrowing gaps

Between 2022 and 2025, socio-economic gaps in performance narrowed across OECD countries. That sounds like progress. It is not.

The OECD explains: narrowing “largely reflected declines among advantaged students rather than improvements among disadvantaged students.” Equity improved because the top came down, not because the bottom came up.

There is a related detail in the engagement data that I think parents should know. Students reported less family support compared to 2022 in almost all countries and economies, “particularly among boys and disadvantaged students.” Fewer students reported that their parents asked what they had done at school that day. The OECD is careful to note this trend appeared in countries with big reading declines and in countries without them, so it does not explain the score drop. But it is one of the few variables on this list that a parent controls completely.

What to actually do at home if your child is at risk

Find out which skill, not which grade

“Struggling in math” is not diagnostic. Below Level 2 in PISA math means a student cannot reliably interpret a simple situation mathematically. Ask the school which specific skills are missing: number sense, multi-step problems, reading the question. The intervention differs for each. Our piece on why kids struggle with math covers how to separate anxiety from a learning difference.

Get a tutor before you get a tablet

The tutoring meta-analysis is one of the most reliable findings in education research: 0.37 SD, stronger for trained tutors, stronger during the school day. If your child is near the baseline threshold, small-group or one-on-one instruction has better evidence than any app.

Ask the one question that kept declining

“What did you do at school today?” Fewer students reported being asked this in 2025 than in 2022. It costs nothing and is the family-support variable PISA actually measured.

Target mid-performers with mindset work, not strugglers

Here is a detail most coverage misses. The growth-mindset score gap is largest among mid-performers (Levels 2 to 4), and previous PISA analyses found the relationships “often non-significant amongst low performers.” A child who is genuinely below baseline needs skills, not a poster about effort.

Protect the disciplinary climate you can control

You cannot fix your child’s classroom. You can make your house’s homework hour quiet, predictable, and phone-free. The systems that improved most in disciplinary climate also tended to hold science performance steady.

What not to do

Do not assume a struggling child is a lazy child. Perseverance fell across 32 to 40 systems since 2022 and declined alongside curiosity, which suggests something about learning environments, not a generation-wide character defect. And do not assume family income determines the outcome. Socio-economic status explains 12% of within-country variation in science. That is real and it is not fate.

What to Watch For Over the Next 3 Months

• Week 4: You know which specific skills your child is missing, in words a teacher used, not just a letter grade.
• Month 2 red flags: Homework completed fast and wrong, or avoidance of any task with a long text. Both are the pattern PISA identified in the declining group.
• Month 3 self-check: If you started tutoring or small-group help, is there a measurable change in one specific skill? If nothing moved, change the intervention, not the child.

Frequently Asked Questions

What counts as a low performer in PISA 2025?

A student scoring below Level 2, PISA’s baseline proficiency, on a subject scale. The headline figure that 20% of OECD 15-year-olds are low performers counts students below Level 2 in science, reading, and mathematics simultaneously, which is a stricter bar than any single subject.

Which countries reduced their share of low performers?

Twelve systems reduced low performers over the past decade. Only five, Montenegro, Qatar, the Slovak Republic, Türkiye, and the United Arab Emirates, reduced low performers while also increasing top performers.

Does more school spending reduce low performance?

Not reliably. The OECD reports that national wealth accounts for about half the variation between countries’ performance, and offers Luxembourg as the counterexample: the world’s biggest spender per student at USD 288,521, with lower science scores than Ireland, France, and Italy.

Does tutoring actually help struggling students?

Yes, and it has the strongest evidence base of anything on this list. The Nickow, Oreopoulos, and Quan meta-analysis of randomized trials found a pooled effect of 0.37 standard deviations, with larger effects for teacher and paraprofessional tutors and for programs run during the school day.

Is a growth mindset enough to move a struggling student?

Probably not on its own. PISA 2025 finds a 28-point science gap associated with growth mindset after controls, but earlier PISA analysis shows the gap is largest for mid-performers and often not statistically significant among low performers. Skills first, framing second.

---

About the author

Ricky Flores is the founder of HiWave Makers and an electrical engineer with 15+ years of experience building consumer technology at Apple, Samsung, and Texas Instruments. He writes about how kids learn to build, think, and create in a tech-saturated world. Read more at hiwavemakers.com.

---

Sources

1. OECD. (2026). “PISA 2025 Results (Volume I): Future-Ready Students,” Preface, Executive Summary, and Chapter 2. OECD Publishing. https://www.oecd.org/en/publications/pisa-2025-results-volume-i_73451bc5-en.html
2. OECD. (2026). “PISA 2025: Students’ reading and mathematics performance declined sharply across the OECD.” OECD press release, September 8, 2026. https://www.oecd.org/en/about/news/press-releases/2026/09/pisa-2025-students-reading-and-mathematics-performance-declined-sharply-across-the-oecd.html
3. Nickow, A., Oreopoulos, P., & Quan, V. (2020). “The Impressive Effects of Tutoring on PreK-12 Learning: A Systematic Review and Meta-Analysis of the Experimental Evidence.” NBER Working Paper 27476. https://www.nber.org/papers/w27476
4. OECD. (2025). “TALIS 2024 Results.” OECD Publishing. https://www.oecd.org/en/about/programmes/talis.html
5. National Center for Education Statistics. (2026). “Program for International Student Assessment (PISA).” NCES. https://nces.ed.gov/surveys/pisa/
6. Institute of Education Sciences. “High Quality Tutoring: An Evidence-Based Strategy to Tackle Learning Loss.” IES, U.S. Department of Education. https://ies.ed.gov/ncee/edlabs/regions/west/Blogs/Details/34
7. Barnum, M., & Meltzer, E. (2026, September 8). “PISA results 2025 show U.S. gaining as other countries decline.” Chalkbeat. https://www.chalkbeat.org/2026/09/08/pisa-results-2025-show-us-gaining-as-other-countries-decline/