Peer Effects in European Primary Schools: Evidence from the Progress in International Reading Literacy Study

TLDR

The observed variation could be explained either by selection into schools or by measurement error in the peer variable. The study estimates peer effects for fourth graders in six European countries. Peer effects are estimated using variation across classes within schools, which the authors argue are formed roughly randomly. Within‑school estimates are much reduced compared to standard OLS, but after correcting for measurement error they are close to the original OLS, indicating a modestly large peer effect, that measurement error is important, and that selection plays little role.

Abstract

We estimate peer effects for fourth graders in six European countries. The identification relies on variation across classes within schools, which we argue are formed roughly randomly. The estimates are much reduced within schools compared to the standard ordinary least squares (OLS) results. This could be explained either by selection into schools or by measurement error in the peer variable. Correcting for measurement error, we find within‐school estimates close to the original OLS estimates. Our results suggest that the peer effect is modestly large, measurement error is important in our survey data, and selection plays little role in biasing peer effects estimates.