Equal scores, unequal advice: class rank shapes Dutch school recommendations

Pupils who score exactly the same on the Dutch end-of-primary-school test can receive different secondary school recommendations depending on how their classmates perform, a new Erasmus School of Economics study has found.

Children who rank as the top performers in a weaker class are more likely to receive a higher recommendation, such as for vwo, than peers with identical scores who are in stronger-performing classes, according to the research. Conversely, pupils at the bottom of a stronger class face a greater chance of a lower recommendation, such as vmbo.

The study, led by Max Coveney, an assistant professor at the Erasmus School of Economics, analyzed data from Statistics Netherlands (CBS) covering about 460,000 pupils from 200,000 families between 2006 and 2021. It examined end-of-primary test scores across more than 5,000 elementary schools, along with the school recommendations given, pupils’ background characteristics, and their later educational outcomes.

Coveney expressed surprise at the strength of the effect. “I had not expected that the position of a child in the class would be so decisive for the advice that it gets,” he told Trouw. As an Australian, Coveney comes from an education system where selection by level occurs at a later age rather than at the end of primary school.

The influence appears strongest among the most experienced teachers, the study found. Teachers may not always be aware of the factors driving their recommendations. “It could be that they unconsciously use the rank order of a child in the class as a kind of benchmark,” Coveney told Trouw.

Another possible explanation involves pupil behavior. “It could also be that the better-performing children in the class radiate more self-confidence, on which teachers base their judgment,” he said. “Then they let themselves be consciously influenced by the behavior of pupils, what confidence in their own abilities is, and how they behave further in the class.”

Coveney believes that informing teachers about their conscious or unconscious biases can help reverse some effects.

The findings align with earlier research showing that certain groups of pupils are systematically underestimated by their teachers. These include children from families with low economic status, pupils from rural areas, and students whose parents did not complete higher education (hbo or wo). Girls are underestimated more often than boys.