Better than all the Rest: the Effect of Ordinal Rank on Educational Performance during Primary School

In this paper we study the effect of a student’s ordinal rank in grade 2 of primary school, at the age of 5, on educational performance in grade 8. We exploit the idiosyncratic variation in the ability distributions that exists both within and across schools and cohorts to estimate a causal effect. Using longitudinal data from primary schools in the Netherlands we show that, after controlling for ability, ordinal rank has a positive effect on both test scores and school advice at a very young age. A one standard deviation increase in rank for the subject Dutch increases test scores with 0.083 standard deviations and school advice with 0.123 standard deviations. The results are stronger for mathematics: a one standard deviation increase in rank increases test scores with 0.226 standard deviations and school advice with 0.256 standard deviations. We find support for three possible mechanisms behind these effects: students with a higher ordinal rank have more confidence, have a better work attitude and receive more extra material from their teacher.