Francesc Pedró: "I have worked for ten years at the OECD and what has happened with PISA in Catalonia is unprecedented"

BarcelonaThis Friday, the Government announced that an error in the sample of selected students has resulted in Catalonia's PISA test results being invalidated for comparison with other territories or previous years. The error by Education means that the sample is not representative and it will not be possible to know if the Catalan educational level has improved or worsened. L'ARA interviews Francesc Pedró, who has worked for a decade with the OECD in the educational field. He is also a professor at Pompeu Fabra University (UPF) and co-director of the 2026 Yearbook of Education in Catalonia from the Equitat.org Foundation.

How does he assess what has happened?

— Well, it's difficult to assess, because it's truly an... unprecedented situation. Having worked for almost 10 years at the OECD, I have no recollection of a similar situation that ended with the withdrawal of the sample. I don't know how the exclusion could have been over 23% of the sample. This is unheard of. The Department of Education will have to explain it sooner or later. I know they have started an investigation, but it's really hard to understand. I do believe that the honesty with which the situation has been handled should be applauded.

Are the OECD criteria for deciding which students are excluded clear?

— The criteria are published and are clear. There is a technical memory that is improved year after year and that specifies, among other things, how the sample should be taken, exceptions, etc.

23% of the students in the sample have been exempted. What percentage of Catalan students do you think have the characteristics to be exempted from PISA?

— In Catalonia, the group of students with classic special educational needs, which are behavioral, cognitive, or even physical [the criteria considered by the OECD], I don't think it reaches 4%. When the OECD says that the sample can exclude a maximum of 5% it is because it assumes that if anything happened that prevented the result from being reliable enough, what would be done is to exclude that student body. But in Catalonia as a whole, we have at most 4% of students like that. Therefore, setting the threshold at 5% is a more than reasonable measure by the OECD.

What exactly does it imply that the OECD considers the final sample to be incorrect?

— In Spain, as in other territories with different jurisdictions, there is a test for the entire State that includes students from Catalonia and from each and every one of the autonomous communities. Furthermore, for some years now, practically all autonomies have their own test. The one that has been withdrawn is the test that should give us results exclusively from Catalonia, not Catalonia's participation in the State test.

But if the sample is not representative for Catalonia, how is it representative for calculating the Spanish average?

— They are two different levels. You can be representative at the state level if you include 100 students for each autonomous community in the state sample. If this is done well, that's it. It's as if you were taking a sample of people sunbathing on Spanish beaches. If I am on a Catalan beach and when conducting the survey people from this beach are left out, the sample for my beach would be incorrect. On the other hand, if the survey is for all Spanish beaches, the fact that there are fewer people on mine is not representative enough to affect the sample of all people sunbathing in Spain.

And in figures, how would I explain it?

— To have the Spanish sample, there are only 150, 200, or 300 Catalans, and that's enough for us. But to have a representative sample of the diversity of Catalonia, 2,500 are needed. If you exclude 500 of these 2,500 because you believe they do not meet the criteria, the sample becomes invalidated. You do not have a sufficient number of students who have been tested. It is curious, however, that they did not know this before.

Will the results we know in September be of any use?

— I don't know. I would say that if they are not representative, they are useless. We cannot make the falsehood of saying that they are not quite representative. Either they are representative or they are not. There are scientific criteria that determine whether a sample is representative or not. If we have not managed to obtain the requested sample, these results are not representative.

What did you expect from these results?

— Frankly, the results could not be much different from what they had been three years earlier. Trends in PISA assessments usually show upward and downward lines, but not radical changes in a three-year cycle. If a country is in mediocrity, it is most likely to remain in mediocrity three years later. Therefore, let's say we could not expect very good results. I wish I were wrong, but what is certain is that regardless of what had happened with these results, national tests already showed the progressive deterioration of students' learning.

Since PISA 2022, have we as a system done anything better?

— Everyone is very well-intentioned, but I insist that the serious problem our system has is that in the last two decades the profile of students has become increasingly complex. We have tools that allow us to check to what extent this diversity is becoming more evident. The truth is that with the level of investment in the educational system, results cannot be expected to improve. Because having a more complex student body requires greater attention. Greater attention requires a larger volume of teaching staff and specialists in special educational needs, the integrators... Deep down, you have more needs and you have to accompany them with more resources. We have investment levels that are below the Spanish average.

Will we notice an improvement with the increase in budget for Education that has been approved?

— The good news is that the budgets that have been approved mean a very significant increase in resources for the educational system. The question now is how these additional resources are being used and if they are precisely aimed at favoring students in disadvantaged situations having more means. And that the centers that are in a situation of greater difficulty have more resources. If you invest more in those who need it most, you manage to raise the country's average. Until now, the lack of resources has been largely supplemented by the dedication and volunteerism of teachers. It is normal that the glass has overflowed and they complain.