Id Foundation

The results of the PISA 2022 survey, involving 700,000 15-year-olds in 81 OECD member and partner countries (8,000 in France), are now known and have been widely commented on: an unprecedented drop in student performance across the OECD as a whole, compared with the results obtained in the 2018 survey. The average drop is 15 points in mathematics and 10 points in reading comprehension, bearing in mind that a full year of education at age 15 is equivalent to 20 points. This drop, according to the OECD report, is only partly attributable to the COVID pandemic that occurred between the last report and the previous one, and this part is probably not major. In addition, the distraction caused by digital devices, both in and out of class, is highlighted as being correlated with poorer results. Finally, the report notes a decline in parental involvement in pupils’ learning, an involvement that is statistically favorable to pupil performance. On the other hand, the report’s preface does not mention the consequences of immigration as an explanation for the collapse in the level of the pupils tested. Yet a careful reading of the results reveals a significant and probably underestimated impact of immigration on pupil performance, specifically in Europe, and even more so in the north of the European Union, a region marked by a strong growth in the proportion of pupils with a migrant background (see table in appendix). The following observations focus on the following nine countries: France, Belgium, Netherlands, Germany, Slovenia, Austria, Denmark, Sweden and Finland.

In its “summary” for decision-makers and the press, the OECD considers, as in its previous reports, that immigration is a factor of inequity in student performance, on a par with socio-economic background and gender, since native students achieve better results in mathematics than students with a migrant background in OECD countries (29 points higher on average, equivalent to one and a half years of schooling). These results confirm a significant gap already observed in previous reports: for example, 43 points in 2015 for science results. However, this 29-point gap is qualified by the OECD and reduced to 5 points, “after controlling for socio-economic background and language spoken in the family.” This nuance, which might lead one to believe that the impact of immigration on school performance in the OECD is negligible, calls for three observations which tend, on the contrary, to suggest that the impact of immigration is underestimated in the European Union.

The subject studied

In 2022, the OECD has chosen to further develop the comparison between immigrant and native pupils in mathematics, particularly in its summary. However, the performance gap is greater in science and reading comprehension, which is not surprising given that results in these two areas depend more on the level of practice (reading and writing) of the host country’s language, which is not the language spoken at home by the majority of pupils from a migrant background (see below). In 2022, there will be an unfavorable gap of 39 points in reading comprehension among pupils from a migrant background in the OECD (versus 29 points in mathematics). In 2015, the gap was 43 points in science and 37 points in mathematics.

Controlling for “socio-economic background and language spoken at home

This statistical correction by the OECD tends to artificially cancel out the effect of immigration, by cancelling out parameters that are nonetheless inseparable from the reality of immigration. It is, of course, mainly in immigrant families that a foreign language is spoken most often, and it’s not surprising that this has a negative influence on school performance.

As for socio-economic background, its unfavorable correlation with immigration is particularly strong in the nine countries studied. The proportion of immigrant pupils considered disadvantaged is around 50%, compared with around 20% of native pupils. This gap is double the OECD average: 37% vs. 22%. This trend was already present in previous reports. In France, the OECD points out that 48% of immigrant pupils come from a disadvantaged background, compared with 20% of native pupils.

With regard to the language spoken at home, the OECD observed in 2022 that the proportion of immigrant pupils speaking a language other than that of the host country is rising steadily. For example, 58% of immigrant pupils surveyed spoke a different language at home, compared with 11% of all pupils, whether immigrant or not. Once again, these discrepancies are accentuated when we look at our nine European Union countries. The seven countries with the most pronounced linguistic differences are European. In Slovenia, Finland, Sweden and Austria, over 75% of immigrant pupils do not speak the language of their host country at home. What’s more, in the nine countries studied, this proportion has never been so high.

To cancel out these two “biases” is to analyze a virtual immigration, where there is no gap between socio-economic background and society as a whole, and where the language of the host country is spoken at home.

A regional disparity highly unfavorable to the European Union

The PISA 2022 report shows an unfavorable gap in mathematical performance of 15 points (equivalent to three quarters of a year’s learning) for students from a migrant background in the OECD, after controlling for socio-economic background. This gap is an average that in reality conceals major regional disparities and tends to mask a much wider gap in Europe. In fact, in many countries, the impact of immigration is positive. This is particularly the case in Qatar and the United Arab Emirates (respectively +66 points and +88 points after controlling for socio-economic background), and to a lesser extent in the United States, Canada and Australia. But in all nine countries studied, the gap is still negative and larger than the OECD average. After controlling for socio-economic background, it reaches 27 points in the Netherlands, 28 points in Denmark, 29 points in Slovenia, and even 42 points in Finland. In countries with a high proportion of immigrant pupils (over 20%), the gap is just as wide: 25 points in Austria and Belgium, 32 points in Germany and 34 points in Sweden. There is therefore a very wide gap in the countries of Northern Europe.

It should be remembered that these already significant gaps (30 points correspond to a year and a half of schooling) are quantified once a correction for socio-economic background has been made, which shows that they cannot be explained by socio-economic bias alone. If we look at the raw differences, we see much higher values. For first-generation immigrant pupils, they reach, negatively, 60 points in France (3 years), 76 points in Sweden, 77 points in the Netherlands (almost 4 years), and 97 points in Germany (almost 5 years of schooling), a gap that is only 44 points on average in the OECD.

Specific observations for France

When we look more closely at the document devoted to France (Programme for International Student Assessment (PISA): Main results for France from PISA 2022), we see that France stands out for its “very worrying disciplinary climate”, according to the OECD (6,770 students took part in the survey). 29% of French students declared that they “could not work well during most or all of their math lessons” (the OECD average is 23% of students), while 50% of students declared that “there was noise and disorder in most or all lessons” (the OECD average is 30%). These results confirm an anomaly already noted in previous reports: in 2018 the OECD pointed out that, in the OECD, only Argentina and Brazil had a disciplinary climate as unfavorable as France’s. The OECD also points out that the disciplinary climate is better in advantaged schools than in disadvantaged ones. Teachers are well aware of this fact, as the harsh disciplinary climate degrades their working conditions. So much so that in the most disadvantaged schools, classified as REP+, the specific bonuses granted by the State to teachers at the start of their careers amount to almost 20% of their salary. This is the price we have to pay to provide a public service in these establishments. In the department of Seine-Saint-Denis, this bonus is reinforced by a €10,000 loyalty bonus for teachers working there for five consecutive years. Seine-Saint-Denis is the department with the largest network of priority education establishments in France (587 REP and REP+ establishments at the start of the 2022 school year), and also the one with the highest proportion of immigrant pupils. Finally, the feeling of insecurity among students is higher in France than in the average OECD country. 12% of students “said they had seen a student with a knife or gun at the school in the month preceding the survey”, a figure higher than the OECD average, compared with 5% in the USA.

Conclusion

The impact of immigration on the overall performance of pupils in the European Union is therefore very significant, particularly in northern Europe where the weight of immigration is high. If we consider the raw gaps in mathematics (before corrections linked to environment and language), in Austria, Germany, Sweden and Belgium, first- and second-generation immigrant pupils score almost 60 points lower than native pupils – the equivalent of three years of schooling – and represent more than 20% of the pupils tested. In its 2015 report, which further analyzed science results, this negative difference reached almost 70 points in Denmark, Germany, Sweden, Austria and Slovenia, the five OECD countries in which this gap was widest. In reading comprehension, the gaps in 2022 are also greater than in mathematics, exceeding 60 points in Austria, Germany, Denmark, Sweden and Finland (and almost 90 points in Finland, the largest gap of the 81 countries tested). This suggests that immigration is a partial but not insignificant explanation for the particularly sharp drop in student performance between 2018 and 2022 in some traditionally high-performing countries where the weight of immigration has increased significantly over the years: in the Netherlands, scores have fallen by 27 points in mathematics and 26 points in reading comprehension, in Finland by 23 points in mathematics and 30 points in reading comprehension, in Germany by 25 points in mathematics and 18 points in reading comprehension, in Sweden and France by 21 points in mathematics and 19 points in reading comprehension. These declines are very worrying, as they represent a year to a year and a half of lost learning over a four-year period.

Appendix: Percentage of students with a migrant background (figures from PISA reports)