Peer effects: they exist but they're not very big

One reason parents try and get their kids into 'good' schools is that they have better teachers, facilities and so on. Another is that the other students are also high achievers and this is believed to feed into their own children's achievement—via less disruptiveness, an environment more conducive to scholarly activity, and so on. A paper newly published in the Oxford Bulletin of Economics and Statistics tests the size of these effects on achievement by looking at the random component of sorting that occurs when most British children transition from primary to secondary school at age 11.

"Peer Effects: Evidence from Secondary School Transition in England" (up-to-date gated version, full working paper pdf), by Stephen Gibbons and Shqiponja Telhaj, finds that although having brighter peers raises someone's grades a bit, the effect size is very small.

Our general finding is that school-level peer effects exist, but they are small in magnitude: a one standard deviation increase in the mean ks2 primary school scores of secondary schoolmates is associated with a 0.03 standard deviation increase in student achievement in secondary school ks3 achievement.

These peer effects originate in characteristics of secondary school peers that were already evident in their achievements at age 7, and family background issues such as low income and English being second language, rather than academic progression during the later years of primary schooling preceding secondary school entry.

This finding suggests a rather limited role for peer effects in amplifying the effects of educational interventions (e.g. social multiplier effects as in Glaeser Sacerdote and Scheinkman 2003), unless these interventions occur very early on in life. Our results show only limited heterogeneity across student demographic types.

But the paper does go on to say that because school has very little impact on student outcomes, we should probably see this as a relatively large effect in that context. And that peer groups might matter for lots of other things besides achievement ("physical safety, emotional security, familiarity, life-time friendship networks, or simply exclusivity") so parents aren't necessarily crazy to aim for 'better' peers for their kids.