Lebanon’s 15-year-old students performed well below international averages in reading, mathematics and science in the latest OECD assessment, highlighting persistent weaknesses in basic skills amid years of economic turmoil, disrupted schooling and conflict.

The findings, released by the Organization for Economic Cooperation and Development (OECD) as part of its 2025 Programme for International Student Assessment (PISA), placed Lebanon among the lower-performing education systems covered by the study.

Lebanese students scored an average of 331 points in reading, 377 in mathematics and 372 in science, all well below the OECD averages.

Lebanon also performed below the OECD average in computational problem-solving, an assessment designed to measure students’ ability to solve problems in a digital environment.

PISA assesses not only what 15-year-olds know but how well they can apply their knowledge and skills to complex, real-world problems, think critically and communicate effectively. Questionnaires completed by students and school principals also provide information about the environments in which students learn.

Lebanon took part in PISA for the first time in 2015, allowing its results to be compared with other education systems and tracked over time.

Compared with 2015, Lebanon’s average scores were 20 points lower in mathematics, 19 points lower in reading and 15 points lower in science. Results in all three subjects were also lower in 2025 than in 2018.

But the OECD cautioned against interpreting the decline as straightforward evidence that the quality of learning has deteriorated across the board.

Expansion of secondary education changes the picture

Lebanon has significantly expanded access to secondary education over the past decade, bringing more 15-year-olds from previously marginalized groups into the school system.

The number of Lebanese 15-year-olds eligible to sit the PISA test increased significantly between 2015 and 2025 despite a broadly stable population of 15-year-olds, suggesting a substantial expansion in secondary-school enrollment.

The OECD said this change in the composition of the student population can entirely explain the apparent decline in national PISA scores.

Among the more advantaged students who were already represented in the 2015 assessment and whose circumstances were not changed by the expansion of education, science performance improved, while reading and mathematics remained broadly close to their 2015 levels.

The picture has also changed among students at different performance levels. Between 2018 and 2025, the gap between Lebanon’s highest-performing students and its weakest-performing students narrowed in mathematics and reading, while there was no significant change in science.

Nevertheless, compared with 2015, the share of low-performing students scoring below Level 2 increased by nine percentage points in mathematics and eight percentage points in reading. There was no significant change in the proportion of low-performing students in science.

Basic reading proficiency remains elusive

The reading results are particularly concerning.

Only 22% of students in Lebanon reached at least Level 2 in reading, the minimum proficiency threshold used by the OECD, compared with 69% across OECD countries.

At this level, students should be able to identify the main idea in a moderately long text, find information using explicit, though sometimes complex, criteria, and reflect on the purpose and form of a text when explicitly directed to do so.

The share of students reaching the highest reading levels was also extremely small. Almost no Lebanese students reached Level 5 or higher, compared with 6% across OECD countries.

Students performing at those levels are capable of understanding lengthy texts, dealing with abstract or counterintuitive concepts and distinguishing between fact and opinion using implicit clues relating to the content or source of information.

The results do not mean that most Lebanese students cannot read. Rather, they show that a large majority struggle to reach the level of comprehension, interpretation and evaluation expected of 15-year-olds in the PISA framework.

Weak performance in mathematics and science

Mathematics performance was similarly weak.

Only 31% of Lebanese students achieved at least Level 2 in mathematics, compared with an OECD average of 65%.

Students at that level can recognize and interpret, without direct instructions, how a simple real-world situation can be represented mathematically, such as comparing distances between two routes or converting prices into another currency.

Only 1% of Lebanese students reached Level 5 or 6 in mathematics, compared with 8% across OECD countries.

At those highest levels, students can model complex situations mathematically and select, compare and evaluate appropriate strategies for solving difficult problems.

In science, 34% of Lebanese students reached at least Level 2, compared with 74% across OECD countries.

At the baseline level, students should be able to recognize the correct explanation for familiar scientific phenomena and use scientific knowledge to determine, in simple cases, whether a conclusion is supported by data.

Only 1% of Lebanese students were top science performers at Level 5 or 6, compared with 7% across OECD countries.

Students at those levels can apply scientific knowledge creatively and independently to a wide range of situations, including unfamiliar ones.

Socioeconomic divide

The results also show a significant performance gap between students from different socioeconomic backgrounds.

Thirty-seven percent of Lebanese students who took the assessment were in the top international quartile of socioeconomic status, making them among the most advantaged students participating in PISA 2025. Their average science score was 420 points.

Students in Lebanon from similarly advantaged backgrounds nevertheless tended to perform significantly worse than students with comparable socioeconomic backgrounds in high-performing systems such as Singapore, Chinese Taipei and Macao.

Within Lebanon, students in the top 25% of socioeconomic status outperformed those in the bottom 25% by 74 points in science. That compares with an average gap of 85 points across OECD countries.

The gap remained broadly stable in Lebanon between 2015 and 2025, while it narrowed on average across OECD countries.

Socioeconomic status accounted for 8% of the variation in science performance in Lebanon, compared with 12% across OECD countries.

There was, however, a notable group of academically resilient students. Some 17% of disadvantaged Lebanese students performed in the top quarter of science performance within the country, compared with an OECD average of 12%.

Students show mixed attitudes toward learning

The assessment found that Lebanese students generally reported positive attitudes toward some aspects of learning, although their results were mixed.

Some 64% said they were curious about many different things, compared with 73% across OECD countries, while 58% said they put in additional effort when schoolwork becomes difficult, close to the OECD average of 60%.

But 32% said they agreed or strongly agreed that school had been a waste of time, compared with 24% across OECD countries.

The OECD found that curiosity, perseverance and students’ perception of the value of school were all associated with stronger science performance.

After accounting for students’ and schools’ socioeconomic profiles, a one-unit increase in the curiosity index was associated with a 29-point difference in science performance in Lebanon, compared with 19 points across OECD countries.

A one-unit increase in perseverance was associated with a 21-point difference in Lebanon, compared with 13 points on average across OECD countries.

Students who placed greater value on school also performed better. Those who did not view school as a waste of time scored 31 points higher in science than peers who felt they were wasting their time at school, compared with a 27-point difference across OECD countries.

Only 58% of Lebanese students reported having a growth mindset — the belief that intelligence can be developed through effort, work and feedback — compared with 69% across OECD countries.

The difference in science performance between students with a growth mindset and those with a fixed mindset was particularly large in Lebanon: 45 points, compared with 28 points on average across OECD countries.

Lebanese students also reported relatively strong self-regulation and goal-setting behaviors. About 53% said they often or very often plan their approach to studying, compared with 37% across OECD countries.

Some 57% said they ask themselves questions about study material to check whether they have understood it, compared with 49% across OECD countries.

Help-seeking was less common than the OECD average. Seventy-two percent said they ask teachers for help when they need it, compared with 77% across OECD countries, while 68% said they turn to classmates when they do not fully understand something, compared with 82% on average.

Schools face resource and discipline problems

The assessment found that Lebanese schools reported fewer shortages of teaching staff than the OECD average, but material and infrastructure shortages remained a concern.

Fifteen percent of students were in schools whose principals said instruction was hindered at least to some extent by a lack of teaching staff, compared with 39% across OECD countries.

The corresponding figure for shortages of assisting staff was 25%, compared with an OECD average of 39%.

But 34% of Lebanese students attended schools where principals said a lack of educational materials, such as textbooks, information-technology equipment, library or laboratory materials, hindered instruction.

Another 27% were in schools where shortages of physical infrastructure, including buildings, grounds, heating and cooling systems, lighting or acoustics, were considered an obstacle to teaching.

The disciplinary climate in classrooms was another major concern.

Thirty-one percent of Lebanese students said they could not work well in most or all lessons, compared with 19% across OECD countries.

Forty-four percent said noise and disorder disrupted most or every science lesson, compared with 31% across the OECD, while 41% said their classmates did not listen to what the teacher said, compared with 29% internationally.

The OECD noted that both teacher support and classroom discipline declined in Lebanon between 2015 and 2025, while both improved on average across OECD countries.

Countries and economies where classroom discipline improved generally saw their science performance remain stable or increase compared with 2015.

Despite the challenges, a majority of Lebanese students reported positive support from their science teachers. Seventy-three percent said their teacher shows an interest in every student’s learning in most science lessons, while 76% said teachers continue teaching until students understand and 74% said teachers provide extra help when needed.

Those figures were 76%, 82% and 76%, respectively, in 2015.

Absenteeism and bullying

School attendance is another area of concern.

About 37% of Lebanese students said they had skipped at least one day of school during the two weeks before the PISA test, compared with 22% across OECD countries.

Thirty-six percent said they had skipped at least one class, compared with an OECD average of 24%.

Nearly half, 49%, reported arriving late for school, roughly in line with the OECD average of 50%.

Bullying was also more prevalent than in OECD countries overall.

Thirty-three percent of Lebanese students said they had been victims of at least one form of bullying a few times a month or more in 2025, compared with 20% across OECD countries.

Cyberbullying was less common than other forms of bullying, but its academic impact was substantial.

Thirteen percent of Lebanese students said upsetting information about them had been published online without their consent at least a few times a month during the year before the test, compared with only 3% across OECD countries.

Those students scored an average of 78 points lower in science than their peers.

Parents also remained involved in their children’s education, although not all forms of engagement were equally common. Fifty-two percent of Lebanese students said their parents or guardians asked them at least once a week what they had done at school, while 40% said they had weekly discussions with their parents about problems they might be experiencing at school.

Digital devices and AI reshape learning

Lebanese students reported extensive use of digital devices at school.

They spent an average of 1.5 hours a day using digital devices for learning activities, compared with 1.7 hours across OECD countries.

But they also spent an average of 2.1 hours a day using such devices for leisure at school, almost twice the OECD average of 1.1 hours.

Twenty-seven percent of Lebanese students said their peers were often distracted by digital devices during most or all science lessons, broadly comparable to the OECD average of 28%.

But the academic penalty associated with digital distraction was much larger in Lebanon. Students who reported such distraction scored 47 points lower in science than those who did not, after accounting for socioeconomic differences. The corresponding gap across OECD countries was 11 points.

Cellphone restrictions were more widespread in Lebanon than in the OECD overall. Eighty-five percent of Lebanese students attended schools where cellphone use was banned on school premises, compared with 49% across OECD countries.

Artificial intelligence has also rapidly become part of students’ schoolwork.

Forty-eight percent of Lebanese students said they used AI chatbots to help them learn at least once a week, slightly above the OECD average of 46%.

Thirty-six percent said they used AI to conduct preliminary research on a new topic, compared with 31% across OECD countries.

The same share said they used AI to summarize a text they had to read, compared with 30% internationally, while 34% used it to draft text for writing assignments, compared with 29% across OECD countries.

Only 6% of Lebanese students said they never or almost never used AI chatbots for any of the schoolwork tasks examined by PISA, compared with 14% across OECD countries.

Environmental readiness

The assessment also examined students’ ability to understand environmental issues and their readiness to respond to them.

Only 34% of Lebanese students reached baseline proficiency in environmental science, compared with 74% across OECD countries.

Lebanese students nevertheless expressed relatively strong willingness to engage in environmental action.

Seventy-two percent said they believed they could have a positive impact on the environment, compared with 81% across OECD countries, while 74% believed collective action could address environmental problems, compared with 82% internationally.

Seventy percent said they knew how to work with others to have a positive impact on the environment, slightly above the OECD average of 68%.

Fifty-five percent said they were somewhat or very interested in contributing to environmental protection through their future jobs, compared with 62% across OECD countries.

Environmental engagement was also relatively widespread. Some 73% said they had discussed with friends or family how to have a positive impact on the environment during the previous year, compared with 56% across OECD countries.

Sixty-eight percent said they had made purchasing decisions based on the sustainability of products, compared with 54% internationally.

Results affected by conflict

The findings must also be interpreted in light of Lebanon’s ongoing crises and the limitations of the PISA sample.

Lebanon’s education system has endured years of disruption caused by the economic collapse, the COVID-19 pandemic, the 2020 Beirut port explosion and successive rounds of conflict.

The economic crisis contributed to school closures, teacher strikes and growing numbers of children dropping out of education, while families faced increasing difficulty in covering the cost of schooling.

The latest conflict caused further disruption, with schools closed or operating on reduced schedules and public buildings used to shelter displaced families. The impact was particularly severe in southern Lebanon, where more than 100 schools have been damaged during the conflict.

The OECD said that Lebanon’s PISA 2025 results cover all governorates except two that were most affected by the evolving conflicts in the Middle East. It nevertheless judged the available data to be fit for reporting.

That limitation is significant because students whose education was most severely disrupted by the conflict may not have been fully represented in the assessment.

A total of 7,382 students in 270 schools completed the PISA assessment in Lebanon, representing about 56,700 15-year-olds, or an estimated 90% of the country’s total population in that age group.

The students were aged between 15 years and three months and 16 years and two months at the time of the assessment and had completed at least six years of formal schooling.

PISA 2025 involved more than 760,000 students worldwide, representing about 33 million 15-year-olds in 91 participating countries and economies.

Students completed a two-hour test, with each student assessed in two subjects. They were given different combinations of subjects and a mixture of multiple-choice questions and questions requiring them to construct their own responses.

They also completed a questionnaire lasting about 35 minutes covering their attitudes, beliefs, home environment and school experiences, while school principals provided information about school management and the learning environment.

The 2025 assessment focused primarily on science, with mathematics and reading as minor areas. It also included an assessment of learning in the digital world, including computational problem-solving skills.

The OECD said results for science, mathematics, reading and computational problem solving were released on Sept. 8, 2026, while findings on students’ self-directed learning in digital environments are due in 2027.

For Lebanon, the findings present a complicated picture.

The country’s falling average scores over the past decade are partly explained by the successful expansion of secondary education to students from previously marginalized groups. Among students who were already represented in the system in 2015, science performance improved and reading and mathematics remained broadly stable.

Yet the proportion of students failing to reach minimum proficiency remains dramatically higher than in OECD countries, while absenteeism, bullying, classroom disruption and socioeconomic disparities pose additional challenges.

The assessment ultimately points to an education system that has expanded access but continues to struggle to ensure that students acquire the core skills needed to succeed in further education, employment and adult life.