Spanish Education: Between Political Alarmism and the Reality of PISA

We are a people who love to talk badly about ourselves. On September 8, 2026, the new PISA report on the educational level of fifteen-year-old adolescents—girls and boys around the world—will be published. It is a long and somewhat tangled report, in English, filled with data, 375 pages long. There was no time to read it, but the media reacted immediately on the same day: “A total fiasco.” “Spain hits rock bottom in education.” “The worst assessment ever for the Spanish education system.” “Resources are lacking everywhere.” “The worst in the Basque Country and the Valencian Community.” “A warning voice amid the great decline.” “Crash in the PISA report.” “Education hits rock bottom.” “Spain achieves its worst result.” “A sharp drop in Spain.” “The students of Spain sink.” “Descents especially savage.” “The magnitude of the debacle.” “The results send chills.” “Plunge.” “Debacle.” “Sinking.” “Systemic failure”… In addition, the President of the Community of Madrid, Isabel Díaz Ayuso, highlights the “struggle” of her Government against the ideologization of teaching. She blames the collapse of the Spanish education system on the Government of Pedro Sánchez, criticizing “the destruction that socialist education laws represent” (sic).

All this is wrong. The PISA report says otherwise. There is a need for a deeper analysis of the report. What matters most in education (and knowledge) is critical thinking. The PISA report has not been read. The commentary in Spain feeds on the pattern of systematic political confrontation that currently characterizes Spanish society. People adopt radical stances, without a serious analysis of the situation. Moreover, a lot of what our young people learn is outside the education system.

“The comments in Spain feed on the model of systematic political polarization in contemporary Spanish society.”

The PISA report is published every three years. This time (2025) it was applied to half the world’s countries: 91 countries. But Russia and India were not included, and China was included in a very limited way. At least the United States is included. Its results are thus partial. It assesses a sample in each country of fifteen-year-olds who attend school (it does not include those who are not in the education system at that age). It primarily evaluates their scientific knowledge, reading comprehension, their mathematics skills, and this time also problem-solving using computers. A subsample measures their English knowledge. Thus, volume two is Connected Through English and volume three deals with Effective Learners in the Digital World.

A conclusion is that the students in the tests perform a reading that is hasty (precipitous or cursory) which consists of skimming a text or just going over it and quickly drawing a conclusion that ends up being false and mistaken. That is precisely what journalists in our country did the day after the PISA report was published. It is impossible to read the 375 pages in print and the hundreds of online tables in twenty-four hours. Simply impossible. I (and I am bilingual) took three full days to read the report and understand it. While the media concluded with the typical response that “Everything is going wrong.” And that resources are lacking. The blame lies with Pedro Sánchez. Also with the bishops. Typical immediate responses in times of tension.

To begin with, Spain’s results exhibit methodological problems that deserve attention. Due to the high exclusions of students in Catalonia and Murcia, the OECD recommends caution in interpreting the scores. Although it regards national results as still comparable, it warns of a possible upward bias. Catalan authorities had excluded 23% of the sample “for various reasons” (never explained). So Spain’s results are included in the report, but with the caveat of doubtful validity; excluding the Catalan responses. Catalonia is a large and very important educational region in Spain. That is, the Spain results are partial and generally false. The results from the Region of Murcia are also inadequate for similar reasons (unmotivated exclusion of 13% of the sample). This is explicitly stated in the PISA report (on page 18). The conclusions of the Spanish media have been hurried, mistaken, and highly unserious. Let us look at everything a little more calmly. Let us precisely follow PISA’s own advice to students during the tests: no hurry.

Qué es PISA

Here Pisa is not a city in the Italian region of Tuscany. PISA is the Program for International Student Assessment. It is developed within the Organisation for Economic Co-operation and Development (OECD), at its headquarters in Paris. PISA reports have been produced since 2000, every three years. The effort is enormous and the OECD’s capacity for analysis is very high. In 2025 more than 760,000 fifteen-year-old students were assessed across 91 countries (countries and economies). In other words, about half of the countries on Earth, but not half of the world’s population. In the final sample, OECD member countries number 38, and 53 more countries are included. In Spain, 30,000 young people were tested with representative samples from the seventeen autonomous communities. Although Catalonia and Murcia failed.

The tests are applied to fifteen-year-old students (up to sixteen years and two months) who are in school and who have completed at least six years of formal schooling. PISA includes three basic tests: in science, reading comprehension, and mathematics. The tests evaluate the knowledge of young people. In 2025 a fourth test on solving problems using computers was included. There is no theoretical minimum or maximum in the scores of the basic tests. They do not come from a census of students, but from a limited but representative sample. It excludes fifteen-year-olds who are not enrolled in any educational institution. It does not discriminate by citizenship or legal status.

A quick—diagonal—reading of this PISA report leads to erroneous and contradictory conclusions. The important thing is to perform a deep reading, not the immediacy of drawing conclusions from limited indicators. One must study the almost four hundred pages of the report and the hundreds of statistical tables with slowness and depth; never for a newspaper headline (which would be false).

“One must study the almost four hundred pages of the report and the hundreds of statistical tables with slow, deep reading; never for a newspaper headline (which would be wrong).”

There is concern—and it is unclear how to interpret it—that, despite increased resources for education in almost every country in the world, scores have remained steady or have fallen. In the last decade (2015-2025) reading scores have declined by 25 points; mathematics by 22 points; and science has remained basically stable. About 20% of students “fail” in the three basic tests. The decline is more visible among the top students (the brightest), not among those who already have poor grades. PISA estimates that 20 points correspond to an entire school year. Perhaps what is most worrying is the erosion of reading skills, since these condition the other tests. Interestingly, those who declined the most were students from wealthier families, not the poorest.

There is particular criticism of students for excessive use of digital distractions and for the misuse of AI (artificial intelligence). Also for the prevalence of bullying (now called bullying). All this occurs inside and outside of school. The habit of reading for fun declines among fifteen-year-olds and rapid digital reading increases. Many students learn to use AI to hand in school assignments, rather than to help think. Teachers insist that what matters is not access to more information, but learning to interpret it. We must foster depth and not speed. It is, therefore, the ability to think about our own thinking. The goal of education is knowing how we know.

The PISA report never takes into account the political regime of the countries. It mixes democracies with autocracies (and even dictatorships). But the political system is essential for the formation of fifteen-year-old students. Perhaps the OECD does not want to engage in political debates about regimes to avoid conflicts. But it could rely on data from Sweden’s V-Dem or the British Economist Intelligence Unit.

The Chinese problem

The OECD must be pleased to have been able to include some of China in the study and in the written PISA report. On the other hand, the Chinese Government must be pleased that China appears at the top of PISA, with Chinese students as the most intelligent, brilliant, and motivated in the world. In reality this is not the case, because what has been achieved is to include only a city and little more, rather than a representative sample of China. In the report it appears as BSJZ (China), where the abbreviation stands for the cities of Beijing and Shanghai and two small eastern coastal territories along the Yangtze River: Jiangsu and Zhejiang. These are not a representative sample of China, nor do they equal an evaluation of the quality of students across China.

The report also separately includes two SAR zones (Special Administrative Regions): Macau and Hong Kong. These are two lightly autonomous (ever less autonomous or independent), non-democratic cities, which are part of the People’s Republic of China. They depend politically on the Chinese Government in Beijing. Thus, it is a trade-off: (1) China appears in PISA and (2) it is placed at number one in the world. This supports the policy of the Chinese Communist Party that democracy is not necessary for a country’s development. It is even better if the political regime is autocratic or dictatorial—like China’s.

“This supports the policy of the Chinese Communist Party that democracy is not necessary for a country to develop.”

There is no doubt that students in China can be more obedient, study more, not criticize the system or the regime, and study for tests diligently. But from a more advanced perspective, democracy is essential for students to form properly and to be able to express their ideas. With this double favor, China also redresses one of the darkest episodes in its history—the Tiananmen massacre on June 4, 1989. In that central square in Beijing, the People’s Army killed hundreds, perhaps thousands, of students who protested, camping in the square beside a statue of Liberty. The demonstration began after the death of Hu Yaobang. To this day that event and the Tiananmen massacre are neither analyzed nor remembered in Chinese schools, not even in Chinese AI. Zero. It is a taboo: it is not talked about, not remembered, and any June demonstration is repressed every year. Not even today is it known how many students died.

Another obvious reason the OECD cannot publish, but which we can suggest, is that the choice of students and schools in China may have been manipulated by the Chinese Government. Surely (no data this year) the worst students were kept at home on test days. China’s push to achieve a high score is impressive. The 56% of students are allegedly at level 5 or 6, which is hard to believe.

Furthermore, China has managed to have Taiwan (now a democratic country with a progressive Government) appear in the PISA report text as “Chinese Taipei,” the Olympic designation. Taipei is a city (the capital), but it refers to Taiwan — formerly Formosa — which is an independent country. For fifty years it was a Japanese colony. It also has the best possible scores. If you exclude the Chinese territories and the city-state of Singapore, Taiwan is really number one in PISA.

The PISA report also includes Singapore, a city-state between Malaysia and Indonesia. Like all city-states, the scores it obtains are not comparable with countries with larger territories, with rural and sometimes remote areas, beyond a single city. It should be added that Singapore also has an ethnically Chinese population (more or less Han). The same can be said of Macau and Hong Kong. There are also problems with Kosovo and the Kurdistan region in Iraq. Dushanbe is a city (1.2 million inhabitants) and capital of Tajikistan (in Central Asia). All these territories (including those in mainland China) should not be mixed with countries.

“As with all city-states, the scores they achieve are not comparable with countries with large territories, rural areas, and sometimes remote regions beyond a single city.”

The report analyzes countries, partial regions, and cities together in its list, without differentiation. It thus speaks literally of “countries and economies.” This phrasing is also used by the World Economic Forum: “economies,” never “countries.” Both exclude Russia (the Russian Federation) in all studies and statistics. We suppose it is because it is an autocracy; but they do not exclude China, which is also one. They even favor it to appear number one in the quality and excellence of its students worldwide. It is not credible that the best education systems in the world are those of China, Macau, and Singapore. Dictatorships fully control students and also manage the application of PISA.

El caso catalán

In all of Spain, about 30,000 fifteen-year-olds were tested for PISA 2025. Yet Spain’s scores are not valid. That is why the storm surrounding the “collective debacle” and the “general underachievement” of Spain in education is misplaced. Catalan school authorities excluded no less than 23% of students in Catalonia. The maximum exclusion recommended by PISA is 5% of those exempt from taking the test for various reasons. That is why Catalonia’s results were not included in PISA. Also in the Region of Murcia the exclusion rose to 13%. PISA did accept those Murcia results, though advising caution in their analysis due to potential bias.

“The Catalan school authorities excluded no fewer than 23% of students in schools in Catalonia.”

The Catalan Education Minister, Esther Niubó, acknowledges the problem: “The people who were responsible for coordinating the tests were not up to it.” A curious euphemism that does not explain what the problem was. This Catalan crisis led to the dismissal of the department’s number two, the secretary general Teresa Sambola, and three disciplinary files were opened. This event has two consequences. First, these test-execution flaws refer only to Spain, which sets a bad precedent. Second, that Spain’s PISA scores must be approached with great caution, and generally the results are partial and not easily comparable. One should not draw sensational conclusions without considering the limitations of these Spanish data.

No es un ‘ranking’, pero casi

The world tilts eastward. If regions, cities, and countries are mixed, the highest positions in PISA cluster in China (two cities and two small coastal regions), Singapore, Japan, South Korea, Macao (a city in China), and Taiwan. Elsewhere, Estonia and the United Kingdom stand out. If you add the fourth dimension—the ability to solve problems using a computer—Chinese enclaves and Singapore stand out. Also Hong Kong (China), Malaysia, Brunei, the Philippines, and Macao.

Since 2012 reading comprehension has been declining worldwide. In reading this is due to the report’s notion of “hasty readers” (i.e., readers who skim). These are readers who race through texts and end up with quick but incorrect answers. This trend has become pronounced since 2018. Mathematical knowledge has also declined since that time.

“These are readers who skim through texts and end up with quick, incorrect answers.”

The developed world does not seem to especially protect the maintenance of quality education. There are countries less wealthy that nevertheless maintain high educational levels. Or in other words: excellent education does not seem to depend (100%) on resources or national wealth. The motivation of youth is important, summarized in PISA as curiosity and perseverance. Also important is identification with the school. This helps explain why girls have progressed more than boys. The recent drop in scores across the three main domains is more due to a decline in education among students from resourceful families than to advances by students from lower- and vulnerable-class families. In addition, boys and low-income families report decreased family support recently.

This is all the more reason to recognize that the system cannot be judged solely by the fortunes of the most advantaged, but must be understood in the context of inclusion.

PISA currently includes four different scores, ranging from 300 to 600 points. There is no theoretical minimum or maximum. It is not a ranking, but one could compute it. The OECD itself has sought to avoid publishing an excellence ranking of youth education because such a ranking would provoke conflicts between countries. But one can calculate the arithmetic mean of the four scores for 2025. Thus, the OECD’s global average would be 476 points. Taiwan, the country with the highest scores, would be 536, i.e., 13% higher than the OECD average. Spain, for example, would sit in the ranking with 471 points, which is 99% of the OECD average. These are country-to-country comparisons. Spain’s position is appropriate, as it almost always sits—not only in education—near the OECD average.

Table 1 shows the scores for the four dimensions (science, reading, mathematics, and computers) for whole countries. It is a selection of countries to better understand Spain’s position in the global context. Spain, for example, achieves an overall score that is 88% of the most advanced country, which is Taiwan. But if we compare Spain with France (which makes a lot of sense, given France’s long-standing emphasis on education), Spain’s overall score in the ranking is more than 99% of France’s score. If we compare Spain with Morocco, Spain exceeds Morocco’s score by 34%.

Up to here, everything seems rational and appropriate. Spain’s situation is fairly good, even correct. It is true that the entire world has fallen slightly in the scores on the table. But that is a global matter of the planet. Spain in this context has not deteriorated much. Rather it maintains with dignity its high position in the PISA ranking. There can be no talk in this context of “fiasco” or of Spain’s “sinking,” much less. This table orders the countries (there are no regions or city-states) by decreasing science scores. It is because PISA 2025’s focus is essentially science. Spain is in 22nd place. This also aligns with rankings of democracy and the human development index. In short, Spain’s position is very good and steady.

We rank countries by homogeneous blocks. Spain should be compared with France (especially France, which has always shown particular attention to education), Italy, Portugal, Greece, and Morocco. These are Southern European countries with many common features. Morocco is Africa, but neighboring Spain. In science, Spain sits in the middle; likewise in reading and mathematics; but in computers it is somewhat higher. All told, everything is balanced and appropriate for the levels of development in our country.

“Spain should be compared with France (especially France, which has always shown particular attention to education), Italy, Portugal, Greece, and Morocco.”

In the table there is also a fifth important indicator: the percentage of top performers. Practically, students who reach the maximum levels—level 5 or level 6—in science, reading, or mathematics. That is, the students who are “bright” or even “bookworms,” academically successful. The OECD average across the 38 countries is that 12% of students are top performers. It is notable that in Taiwan, the world’s top country, the proportion is 35%; in South Korea it is 28%; and in Japan 27%. In Central Europe it ranges from 13% (Belgium, Ireland) to 18% (the United Kingdom or Switzerland). By contrast, southern Europe sees this percentage fall significantly; France, for example, has 10% top-performing students; Spain only 7%; but in Morocco it is only 0.1% of all fifteen-year-old students. We are, then, looking at a indicator that is sensitive to quality and excellence in education and one that we should use more.

The shares of bright students fall sharply in Latin America. Argentina stands out for maintaining a very low education level—contrary to the Argentinian stereotype. It has only 1.2% top students. Mexico only 0.5%. In the other extreme of the continent, Canada and the United States reach 17% and 18% top students. The rest of Africa and Asia have very low proportions. But according to PISA’s report, if one measures the share of top students (the best grades) in the Chinese cities and the Jiangsu and Zhejiang regions, it would be 56%. This is not believable.

Obviously, one of Spain’s most important objectives would be to achieve a higher share of top students (excellent or intelligent). This requires focusing efforts to enlarge that elite. These are the ones who can later pursue a STEM career (they are directly questioned). Because increasing the overall score in science, reading and mathematics for all students is a more complex task. In science, Spain has 99% of the OECD average; in reading, Spain has 98%; and in mathematics it reaches 99%. In computers it already reaches 100% of the OECD average. There is no more to ask of our country. The Spanish education system is efficient and of high quality. In any case, Spain’s current position in the global context is very good, appropriate, stable, and by no means a failure as some yellow press outlets hastily claim.

“Obviously, one of the most important objectives for Spain would be to achieve a higher share of top students (excellent or intelligent).”

On one day, the PISA report had not been read. But there were sensationalist comments. Even less could be analyzed in depth. For example, the day after its presentation in Paris at 9:30 in the morning (i.e., September 9, 2026), the newspaper El País—generally a measured and rational newspaper—opens the front page with a sensational four-column headline: “Fiasco in the PISA Report. Education hits rock bottom. Spain achieves its worst result in the evaluation of fifteen and sixteen-year-olds.” The editorial on page 10 bears the title “Educational fiasco. The PISA Report shows an alarming decline in Spain, a global decline that puts our societies’ future at risk.” It then includes an Ignacio Zafra article that draws attention to Spain’s students’ results as “the worst in history” and “the test warns of a global loss of concentration.” On page 31 there is a triple table with the results of 44 countries in mathematics, science, and reading.

In that table, rightly, it does not include city-states or partially regional areas or other cities. Only countries. We agree; do not mix apples and oranges. In science the best students are in Japan, Estonia and Korea (one must refer to South Korea). In reading, in Japan, Korea (South) and Ireland. In mathematics the best are in Japan, Korea (South) and Estonia. But mysteriously the article omits Taiwan (which in the PISA report is wrongly named “Chinese Taipei”), which is actually the world’s top country. Why is it not included? Then it excludes many other countries that are ahead of Argentina (the last on the newspaper’s list), such as the United Arab Emirates, Vietnam, Mauritius, Uzbekistan, Montenegro, Albania, Qatar, Thailand, Serbia, Mongolia, Georgia, Moldova, Kazakhstan, Malaysia, Saudi Arabia, Azerbaijan, Brazil, Jordan, Peru, Ecuador… and finally includes Argentina. It is not a good idea for country rankings to exclude so many countries. There is a certain assumption that they are “countries of no importance” or incomplete, which entails a negative marginalization.

The news thus reaches the public through sensational yet erroneous and depthless messages. It is necessary to explain that some Asian countries are achieving very high scores, and even less developed countries as well. This is precisely the conclusion of the PISA report’s analysis: more resources do not guarantee better education.

“This is precisely the conclusion of the PISA report’s analysis: more resources do not guarantee better education.”

It is shown that educational systems with more resources and higher per-student spending achieve higher scores in science… but only up to a certain point. After a certain level of resources and spending, increasing those resources does not produce a corresponding increase in the quality of students’ education. At the same time, in systems with fewer resources, progress is possible. It is important to understand the graph on page 55 of the PISA report, on “Notes on science and education spending.” On the horizontal axis, cumulative spending per student over the theoretical duration of studies (in international PPP dollars) is plotted. On the vertical axis, the average science scores appear. At the start of spending, the score grows very rapidly. But after around $100,000 per capita (over the entire career), the growth slows markedly. Spain, Portugal, Italy, and France, for example, lie exactly at that inflection point where more resources no longer yield substantial educational gains. Japan and Taiwan spend resources similarly but obtain better results. From a certain point—precisely where Spain is now—educational progress becomes marginal. Luxembourg, for example, spends far more and still has lower results. The conclusion is clear: beyond a certain point it is not worth investing more money and resources. Other variables must be improved.

All decline

What the PISA report does not explain, nor do commentators, is why the majority of countries in the sample see declines in their reading and mathematics scores over the past years (at least a decade). If the decline were only in Spain, we would have reason to worry; but it declines in every country in the world. It seems to be a systemic pattern, possibly due to several factors. Reading scores decline from 2012 onward; mathematics from 2018 onward. One of the explanatory factors could be the long-term consequences of the COVID-19 pandemic. Another is clearly the appearance and excessive use of screens and digital information by students. Another is simply that students around the world lose the value of effort, becoming lazier, less interested, even more apathetic. It may also relate to the global decline in birth rates. Perhaps it depends on the progressive equality of women (and girls) with respect to men. It is interesting that the recent declines occur more among the best students, not among those from low-income, immigrant, and vulnerable groups. The greatest disparities among students are observed in the United States and Luxembourg. One must distinguish differences between countries, between schools (within the same country), and between students in the same school.

One can venture that the most important cause is that the education system is expanding, enrolling more and more people. It thus includes youths who were previously marginalized or from poor families. The clearest case is immigrants who previously did not enroll or tended to drop out early (early dropouts), even without obtaining a high school diploma. The fact that PISA scores have fallen worldwide could be the consequence of a success: more people who previously did not enroll in schools now do. Another related reason is the inclusion of girls in less developed countries where they were previously excluded from education. Thus global scores have fallen, but inclusion has increased.

“The fact that PISA scores have fallen worldwide may be the result of success: more people who previously did not enroll in schools now do so.”

Finding scapegoats for the systematic decline—such as mobile phones, AI, television, bullying, the decline of reading among youths, or a lack of effort—is easy, but almost certainly wrong. It is a classic victim-blaming trend. We need more research, but it is possible that the average PISA scores fall because more people from lower classes are being included, with poverty, marginalization, vulnerability, or foreign status (immigrants, including undocumented ones). Their integration is genuinely a social achievement. I am not making these ideas up. PISA itself hints at this explanation, but toward the end of the report that apparently no one has read.

In Table 2 I calculate the change that occurs in neighboring countries (Southern Europe) with other reference countries. The table includes change from 2015 to 2025. At the bottom also appear the most recent changes from 2022 to 2025. Indeed, over the decade, the OECD average shows science scores down by 7 points; reading down by as much as 27 points; and mathematics by about 24 points. All of this in ten years. Then I include the results—generally negative, i.e., descending—of Portugal, Spain, France, Italy, Greece, and even Morocco. In the last decade, with the exception of science in Italy (which rises by 5 points), all other data are negative, i.e., scores fall. Greece drops sharply in science, but Morocco surpasses it. In reading, the biggest drops occur in France and Greece. In mathematics, the maximum drop occurs in France. Spain’s situation is balanced—mid-range—so there is no reason to worry excessively. I also include data for the United States (a major reference country), the United Kingdom, and Germany as central European countries; and Finland as the Nordic country that has long stood out for its educational excellence. The reality is that in the last three years, science results hardly change, mathematics fall by 9 points, and reading by about 14 points. It is not excessive. But in OECD countries, 49% of fifteen-year-old boys and girls fail at least one of the three domains (science, reading, or mathematics). In other words, they do not reach level 2.

Table 2. Change in points in the 2015-2025 decade

Change in points between 2022 and 2025

Notes: The two parts of the table present the evolution of the three PISA tests in recent years. Almost all countries have declined (negative sign). The trend is, then, downward, which is the world pattern. We compare Southern Europe with other reference countries.

The Anglophone world is the most stable, while Germany and the United Kingdom exhibit declines more intense even than those in Southern Europe. All evidence suggests that the general decline in this decade is more due to immigration, inclusion of minorities, and vulnerable families. It is, therefore, a very sensitive improvement in the education system: the opposite of what the media claim, which has never read the PISA report. In any case, Spain’s situation is good. The most recent changes (in the last three-year period) follow the same patterns, but with less pronounced declines. In science there are hardly any declines, and in the United Kingdom the score increases. In reading the impact is more noticeable in Spain and France. In mathematics France and Spain stand out more. In the long run Finland’s position has weakened (which is surprising), even in the last three-year period.

In summary, in the last decade the proportion of students below level 2 (which we can assume is a failing grade) has increased by 10%, while the good and bright students (levels 5 and 6) have fallen by only 2%. In any case, PISA scores are now the lowest in twenty-five years. Some people think the “culprits” are the students themselves, with a reduction in their attention and more negative attitudes toward school. Reading is more quickly done. Students tend to answer more quickly, with a risk of error. In schools there is a greater need to assist students. Maintaining order is increasingly difficult. In the last triennium there seems to be more bullying. Teachers spend more time keeping order. Classes are more diverse. There are more immigrant students, including refugees, more students who do not speak the official language, students with special needs, and many more with migration experiences. Some schools concentrate more of these problems.

“There are more immigrant students, including refugees, more students who do not speak the official language, students with special needs, and many more with migration experiences.”

Generally, reading habits of children and their parents have changed. Digital dominates more and more. We must also consider the pandemic’s effects (the 2025 cohort were ten or eleven years old during the COVID-19 pandemic). Some students admit they only read “if there is no other choice.” Online activities are dominant. Reading for pleasure even declines among parents, so young people do not see their parents reading. But students who read fiction and thicker books achieve better grades (and vice versa). It is important to read on paper, not on screen. The opposite effect (negative) is spending too much time on social networks or playing videos. Students keep writing, but mostly in digital media and shorter content. There are fewer conversations with parents (or relatives) at home. But there is no simple explanation for PISA’s education decline. Perhaps the cause is greater diversity in classes and schools; and greater inclusion. There is also greater use of digital tools. It is important to understand these causes to forecast the future of schools and adolescent education.

The PISA report draws attention (pp. 104 and following) to the fact that it is not that the quality of education has declined for all students, but rather that it reflects the expansion of education to more marginalized and segregated populations. The true improvement of inclusive education is thus not captured. Students who were previously excluded are now entering school, pulling down the PISA average, but achieving a clear success for the education system and social cohesion. Fifteen-year-old students who were previously kept out of the education system are now integrated. This trend appears to be universal. The goal is to ensure equal access for vulnerable groups. Inclusion is more important for low-income, vulnerable, and traditionally marginalized boys and girls. The objective is to achieve a high level of equality compatible with better education for everyone. In Spain this is more evident, as the country experiences a considerable increase in immigration.

What must be understood is the impact of family socioeconomic differences and levels of inequality in each country. PISA considers that “a student’s socioeconomic status remains the strongest and most consistent predictor of their academic achievement” (p. 114). The family’s socioeconomic status remains vital. Table 3 includes the percentage of variance in science explained by socioeconomic status. PISA uses the “ESCS” index—the Economic, Social, and Cultural Status. This index explains about 12% of the variance in science scores at national levels. It does not seem a very high proportion. Variations appear somewhat random. Germany shows a high variance (19%), but it is higher in Romania (22%). Countries with high science scores appear with low percentages of variance due to socioeconomic differences: Taiwan 11% and South Korea 8%. Spain’s variance (9%) is half that of Germany (19%), and France has 47% more variance than Spain. Spain, therefore, remains at a middle-lower level of socioeconomic (and cultural) inequalities that feels quite satisfactory and balanced.

“Spain, therefore, remains at a middle-lower level of socioeconomic (and cultural) inequalities that is quite satisfactory and balanced.”

What’s new in this table is that we get scores for the native population (the country’s own nationals) versus youths from immigrant families. The differences are consistent: immigrants achieve science scores 43 points lower than natives, i.e., two complete academic years. That is significant. The table lists the countries where immigrants fare worse or face greater discrimination. From Mongolia with a 111-point gap (in favor of fifteen-year-old natives) down to 53 points in Belgium. Spain also appears, with a 43-point gap between natives and immigrants. This Spanish gap is exactly the OECD average. Again Spain shows differences between immigrants and nationals, but the situation is balanced. The largest gaps are seen in Mongolia, South Korea, and the Philippines, i.e., in the East. Yet Taiwan, which is the top-scoring country in science, has only a 55-point difference. Neighboring France shows a 57-point gap between immigrants and natives, i.e., considerably more than Spain’s 43 points, and close to three full academic years. It is striking that Finland, often exemplary in schooling, maintains large gaps (88 points). Providing education to migrants is a positive feature of the education system, even though it leads to explicit differences between immigrant youths (or youths from immigrant families) and the native population.

Table 3. Social disparities in PISA science scores

(Countries with the largest disparities).

Notes: Here is a selection of those with more than 50-point differences (in science scores) between immigrant and non-immigrant students. The percentage of variance explained by socioeconomic differences is measured with the “ESCS” index.

The case of BSJZ (Beijing, Shanghai, Jiangsu and Zhejiang) in China —which is not representative of China at all— has a variance percentage of 9.7% and a 54-point difference between native and immigrant populations. The largest variance is seen in Zambia (37.8% of variance due to socioeconomic inequalities). Luxembourg’s variance is also significant (21.7%, but with only 21-point difference among migrants). Spain’s case almost exactly matches the OECD average (as is usually the case), but Spain’s variance is a little lower.

Differences against immigrants are measured in points. It is calculated that every 20 points equate to one complete academic year. Thus in Spain, and in OECD average, the 43-point difference for immigrants amounts to two full years of study. In South Korea or the Philippines, they equate to five full academic years of delay.

There are a few countries where immigrants score better than natives: Hong Kong in China, Cyprus, Tajikistan, Montenegro, the United Kingdom (in this case there are no differences), Australia, the United States (with eleven points in favor of immigrants and undocumented students), Brunei, Uruguay, and Ecuador.

Las chicas saben más

Girls are entering the education system and are already the majority in democratic and advanced countries. It is not clear whether women study more, whether they are more curious and persevering (according to PISA indicators), or whether boys are failing more at all levels. In democratic Europe and other democratically advanced countries, women are more frequently in education, obtain better grades, fail less, repeat fewer courses, and drop out less often… although in the end they obtain less well-paid jobs, more part-time work, and with less professional power than men. It is true that girls read more than boys. In Spain, men play more football, retake courses, suspend, and some give up.

According to PISA reports, reading comprehension is basic for all other indicators of excellence. Even for mathematics and science, understanding the question or hypotheses is necessary. In Table 4 you can observe the differences by points between girls and boys in reading comprehension. Except in Rwanda, in all other countries analyzed by PISA, fifteen-year-old girls obtain higher scores in reading comprehension than boys of the same age. On average in the OECD, girls score 30 points higher than boys. That represents about a year and a half of schooling at the same age. Spain is close to this measure with 26 points in favor of girls.

Table 4. Gender differences — girls outperform boys — in reading comprehension

(Includes only countries above the OECD average).

Notes: Here is a selection of countries whose difference in favor of girls exceeds the OECD average (which is 30 points). In all countries girls outperform boys in reading comprehension scores. It is estimated that every 20 points correspond to one full academic year. In Spain there are no gender differences in science, but there is a 14-point advantage for boys in mathematics.

“Except in the case of Rwanda, in all other countries analyzed by PISA, fifteen-year-old girls obtain higher reading comprehension scores than boys of the same age.”

The points advantage for girls (versus boys) orders the table from highest to lowest. Thus the gender gap (in favor of girls) in France is larger than in Spain (with 32 points), similar to Portugal (with 33 points). In Greece it is already 46 points. Differences between countries are small and balanced. It is possible that these differences will widen in the future and even contaminate mathematics results, which are traditionally more male. Across all countries, male students score higher in mathematics than girls. But gender differences are smaller in mathematics than in reading: on average in the OECD they differ by 13 points (in favor of boys). In Spain in mathematics the difference is similar to the average (14 points). In science there are no gender differences in Spain; on average the OECD sees girls ahead by a single point. But there remains a certain gender gap, which is diminishing (the gap).

The presence of women in PISA is an indicator of the progressive incorporation of minorities into educational systems. Spain is a good example with very positive indicators. It is the conjunction of being a democratic, feminist country with a progressive government.

Los nuevos demonios: ‘bullying’ e IA

The current PISA 2025 report concentrates attention on relatively new phenomena that influence knowledge and its applicability to social reality in the education of fifteen-year-olds in the 91 countries surveyed. Among them are bullying, the use of the digital world, specifically AI (artificial intelligence), and concerns about the environment. The analysis is in the report’s second part (which apparently no one has read). A second PISA volume on knowledge and use of the English language will be added to this, but it has not yet been published.

“Among them are ‘bullying’ (or school harassment), the use of the digital world, specifically AI (artificial intelligence), and concerns about the environment.”

The term bullying (the word, not the behavior) is trendy and has already been accepted. It refers to school harassment generally by other students and friends; inside or outside school. The cyberbullying refers to harassment carried out through digital channels, including the internet and social networks. It is not a central topic of the PISA reports, but it remains of growing interest.

In Table 5 I calculate the different cases of bullying for the OECD average and for Spain’s six neighboring southern countries. It is always wise to compare Spain with its surroundings. In this case, school harassment is measured by the frequency “several times a month.” It should be understood that this is a direct self-report by the student, so data reliability is uncertain. The OECD average for bullying reaches 20%, a sizeable and worrying share: it is one fifth of the world’s student body. This harassment is high in Greece (22%) and also in Morocco (21%). In France (18%) it is higher than in Spain (14%). The lowest rate of harassment occurs in Italy (in this sample) at 11%, several times a month. In Germany bullying rises to 31%. In the United Kingdom it is 23%. Among the lower percentages are South Korea (13%) and Taiwan (only 11% total bullying).

Table 5. Students report having experienced harassment at least “several times a month”

Notes: Here is a selection of Southern European countries to understand Spain’s comparative situation. “Students” includes both girls and boys. In Germany, bullying reaches 31%; in the United Kingdom 23%. In Macao (China) it rises to 25% and in Hong Kong (also China) exceeds 22%. Among the lower percentages, besides Italy (11%), are South Korea (13%) and Taiwan (11%).

The gradual levels of bullying are as follows:

• Other students mock me.
• I am threatened by other students.
• I am pushed or bumped by other students.
• Other students spread unpleasant rumors about me.
• Personally identifiable information is published online about me without my consent.

The fifth is what is usually considered cyberbullying. Fortunately, cyberbullying through digital channels (the internet, social networks, WhatsApp, X, Instagram, Facebook, etc.) occurs in only about 3% of students. But even though cyberbullying is infrequent, it can have more negative consequences. It also often occurs in combination with other levels of school harassment. These five levels of bullying are associated with lower PISA test scores. They are also related to absenteeism from class or school without a clear reason.

The most common form of school harassment is that “other students mock me,” with a prevalence of 14%. Slightly less common are students who suffer unpleasant rumors about them, which accounts for 9%. The distribution of harassment is very favorable to Italy. The country must have some secret for achieving such low bullying rates. Spain’s situation is good. When compared with neighboring France, Spain has fewer harassment cases in all modalities. In France, 14% of students admit that other students mock them. In Greece it is 16%. But curiously the lowest rate is in Morocco, with only 8% of students mocked. Yet in all other cases in Morocco students suffer substantially more harassment. Perhaps mocking one another is not considered harassment in Morocco, or it is normalized.

“The most common form of school harassment is that ‘other students mock me,’ with a prevalence of 14%.”

Spain has a medium, even low level of bullying. It does not seem worrying. In parts of China, harassment is even double. There is peer harassment, but the institutional harassment by government, school authorities, or families can be much worse. In non-democratic countries the government’s, school, or family harassment can be enormous. Unfortunately the PISA report does not measure this. It is impossible to achieve excellence and high quality in education, nor to develop a “critical thinking” (which is the primary objective of a good education system) in an autocratic and totalitarian regime. In the future PISA should include the country’s political regime as an indicator in its analyses. A school that hides the truth, distorts its country’s history, and maintains strict obedience cannot be considered in the least as excellent. Moreover, the scores of some autocratic countries are not credible (like China) and may be manipulated at least in the actual sample of students, and even in the responses to the four tests.

The explanation that the general decline in PISA scores is due to contemporary fifteen-year-olds being lazy and intoxicated by the digital world and AI is a typical blaming the victim explanation. The downward trend in scores—modest but global—likely stems from a more inclusive education system that brings in more people. Schools have incorporated minorities, youths who previously did not study or had dropped out by age fifteen after repeating several courses, and now inclusive schools around the world enroll the children of vulnerable, poor, marginalized families, many of whom are divorced or single-parent families (mainly mothers), teen pregnancies, and especially immigrants.

Immigrants retain a 43-point delay in the science variable in PISA. The 43 points measured by PISA correspond to two full years of study. Spain, for example, has significantly increased the proportion of immigrants in schools, to one in five students. But these fifteen-year-olds, even though they lower the average scores, represent a great social success. For a vulnerable, poor, or immigrant family, the only salvation for their children (PISA explicitly states this in the report) is to “study in a well-organized public school, and have a teacher of high quality.” That is why the PISA report recommends that the best teachers go to work in the most problematic schools or those with higher proportions of vulnerable students. Something that illustrates—at least for Italy—the autobiographical tale of Elena Ferrante (pseudonym) in The Wonderful Friend.

It should be stressed that girls (women in general) are making great strides in education systems. To the point that there are already more women than men in universities. In reading—viewed as the fundamental basis of study quality according to PISA—girls in OECD countries systematically obtain higher grades than boys. On average they outperform boys by 30 points, which represents about a year and a half of schooling at the same age. In Spain it is 26 points. But in Finland (often considered an educational model) it is already 46 points, i.e., more than two years’ advantage for girls over boys. The transformation of the educational system and the pursuit of excellence may increasingly come through girls.

“The change in the educational system and the pursuit of excellence may increasingly come through girls.”

It remains to be seen in Spain whether a public system of universal, secular, and democratic schooling (as in neighboring France) can be created. Perhaps the biggest obstacle to a meaningful improvement is overcoming Spain’s dichotomous system of private or quasi-private (fee-based) schools—most of them Catholic—for wealthy families, and free public schools for the poor and immigrants. But this is not easy to achieve, as Spain’s privatization of education continues to rise. That is another issue to analyze.

One must not be lazy, but persevering. The massive effort of PISA (and the OECD) to provide us with these comparative data represents a huge social advance. It is also very costly in time and resources. There remain aspects to improve in the education system. We must prevent two autonomous communities in Spain (Catalonia and Murcia) from failing to carry out their PISA duties properly. We must prevent some countries from lying or whitewashing results; or leaving the worst students at home on PISA test day. But above all we must ensure that PISA reports are read and analyzed with meticulous care, without jumping to erroneous and biased conclusions, typically due to ignorance or political and social tensions. PISA advises students to avoid hasty reading. I would especially recommend this to politicians and their policies (sic). In education, Spain is not a disaster or a shipwreck. We are building a modern education system that is inclusive of our immigrants and minorities, and with a decidedly feminist orientation. Congratulations. As PISA says, “change is possible.”

“We are building a modern education system, which at the same time is inclusive of our immigrants and minorities, and with a decidedly feminist orientation.”

I’ll end now. Thank you for reading this far and for looking carefully at the table data. There is one more thing we must acknowledge. It is the dedication and intelligence of Andreas Schleicher, Director for Education and Skills, at the OECD in Paris. He has devoted at least three decades of his life (a lifetime) to advancing PISA and its reports. The research team at the OECD—with whom I have worked for several months in Paris—is of enormous quality and humanity. Often in science, advances come from specific individuals. In that sense, we owe Schleicher and his entire team immense thanks.