The results of the PISA 2025 international survey show that reading is the skill where Italian fifteen-year-olds are best while the greatest criticality emerges in digital programming. The students of the Bel Paese, on the other hand, are on average OECD in scientific and mathematical subjects and the percentage that reaches at least the level of basic skills is higher than the international average in all three areas analyzed by the survey.
For the first time, no significant differences emerge between boys and girls in average science scores, while girls continue to be better at reading, and boys at mathematics. Social origin also affects the results, albeit with exceptions.
What is the PISA survey and what does it measure
Since 2000, the OECD (Organization for Economic Cooperation and Development) periodically conducts the PISA survey (Program for International Student Assessment) in order to evaluate, through a digital test lasting 2 hours, the skills of fifteen-year-old students from different countries around the world. This age was chosen because in many countries it coincides with the final phase of compulsory education, and was therefore considered the most suitable for understanding how well students know how to use what they learned at school to cope with the problems of adult life.
In 2025, more than 760,000 students from 91 countries and economies participated, representing approximately 33 million 15-year-olds worldwide.
Each edition of PISA delves into one of these areas of competence: science, mathematics and reading.
In 2025 the focus was on science, while mathematics and reading were considered secondary fields. The evaluation of computational problem solving was also introduced for the first time.
But what does it mean, concretely, to be competent in a certain area?
- Science: knowing how to deal with scientific questions and ideas and knowing how to use them to make decisions;
- Mathematics: ability to reason mathematically and use mathematics to solve everyday problems;
- Reading: being able to understand a text, evaluate it and reflect on the information contained to develop new individual and social skills;
- Computational problem solving: knowing how to solve problems through iterative and self-regulated processes, also using digital tools.
Student results in Italy
Reading
Reading is the main strength of Italian students, whose average score is significantly higher than the OECD average. In Italy, 76% of students reach at least level 2, compared to 69% of the OECD average. In practice, this means that more than 3 out of 4 15-year-olds can identify the main idea of a moderate-length text, find information based on explicit criteria, and reflect on the intent or form of a text.
Only 4% of Italian students reach level 5 or above, compared to 6% of the OECD average. In other words, only a small share of students proved capable of understanding long and complex texts, managing abstract or counterintuitive concepts and distinguishing facts from opinions also using implicit clues.
These percentages therefore outline a particular situation: Italy has a share of students with basic skills higher than the OECD average, but a relatively small share of excellent students.
The differences become even more evident when comparing schools. In high schools, 91% of students reach at least level 2, while the percentage drops to 40% in professional institutes and 38% in professional training courses.
Science
As the main evaluation area of the 2025 edition, science occupies a central place in the report. 78% of Italian students reach at least level 2, compared to 74% of the OECD average. This means that more than 3 out of 4 students are able to identify the correct explanation of a known scientific phenomenon and to establish, in simple situations, whether a conclusion can be considered valid based on the data provided.
Even in this case, however, Italian students at the highest levels are relatively few. Only 4% reach level 5 or 6, compared to 7% on the OECD average. The students who reach these levels are in fact only those who manage to apply their scientific knowledge autonomously and creatively to new and unfamiliar situations.
An interesting result then concerns the type of scientific skills possessed. In fact, Italian students show themselves to be particularly competent, especially in the evaluation of research designs, an important skill especially for being able to critically evaluate a scientific experiment and its results.
Mathematics
In mathematics, 69% of Italian students reach at least level 2, compared to 65% of the OECD average. At this level students are able, for example, to compare the distance of two alternative routes or convert a price from one currency to another.
Here too, however, if Italy manages to maintain a good share of students above the minimum threshold, it struggles more in the excellence range: only 7% of Italian students reach levels 5 or 6, compared to 8% of the OECD average. To reach these levels it is necessary to also be able to demonstrate the ability to transform complex situations into mathematical models and to be able to select, compare and evaluate different strategies to solve a problem.
Computational problem solving
One of the less positive results for Italy comes from digital skills. In the new scale dedicated to computational problem solving, the international average score is set at 500 points. Italian students obtain 490 points, a result statistically lower than the international average.
The data is particularly interesting because it shows that knowing how to use digital tools does not necessarily mean knowing how to solve problems digitally. In particular, the greatest difficulty of Italian students seems to concern programming.
Social origin and gender: how much do they matter?
The PISA survey also allows us to analyze the extent to which school results are associated with the socio-economic conditions of students. In Italy, social origin continues to have a significant weight, but the picture is substantially in line with that observed on average in OECD countries and the data show that starting conditions do not automatically determine educational results.
However, gender differences change depending on the subject. In science, for the first time in 2025 the average scores of Italian girls and boys are substantially similar, even if the boys with the best results do better than the best girls. The situation changes in mathematics, where boys achieve better results than girls on average. The gap widens further when considering the highest performing students. In reading, however, girls achieve better results than boys. There is therefore no unique male or female “advantage”, but the differences depend on the competence being measured.
