The latest Pisa survey from December 2019 was a joy message for the Swedish school. The results showed that Swedish students performed better than the OECD average in both mathematics and science and reading comprehension.

In a comment to SVT News, Education Minister Anna Ekström (S) said that the improved results have been made despite the fact that the Swedish school has received a record number of newly arrived students and at the same time wrestled with larger pupils and teacher shortages.

"This is a message of strength, Swedish schools are strong," said the Education Minister.

But the strength message can be based on frayed figures. An examination of Expressen shows, among other things, that a large number of foreign-born students have been wrongly sorted out of the Pisa sample, which violates the OECD's official rules.

"Is extremely serious"

According to the regulations, students who have studied Swedish for a year or more must write the exam. Only those students who have been taught Swedish for less than a year may be exempted.

Expressen, together with Magnus Henrekson, professor at the Institute for Business Research, compiled and analyzed statistics from the National Agency for Education, the Swedish Migration Agency and Statistics Sweden. The review shows that the Swedish Pisa results would probably have looked worse if foreign-born pupils had not been wrongly sorted out.

"This is extremely serious, especially as the Pisa results have in practice been used as a measure both of how the Swedish school stands in comparison with other countries and how it develops over time," Magnus Henrekson told the magazine.

The school system stands firm

The National Agency for Education has previously suspected that Sweden has excluded too many foreign-born pupils from participating in Pisa.

In December, a few days after the Pisa result was presented, the authority published a clarification on its website pointing out that there were too few foreign-born students in the sample.

When the National Agency for Education is confronted with Expressen's own calculations - which shows a greater elimination than the National Agency has concluded - the authority maintains that their calculations are more accurate.

- I maintain that we have to be humble in the face of this situation. Thus, we feel that we have used the best method that we consider to be the most fair, says the Swedish National Agency for Education statistics Anders Auer.