Amine Ouazad, INSEAD
How much do peers `matter’ at work, at school, in the neighborhood? This is a difficult empirical question; people sort themselves into firms, schools, and neighborhoods. For instance, in the classroom, are good pupils good, because they are together or are they together because they are good? School segregation is still strong in the U.S. and in Europe, but it is hard to identify how much segregation shapes performance and inequalities.
First, selection effects and peer effects must be disentangled. Second, pupils influence each other simultaneously, and it is therefore hard to separate the effect of peers’ behavior from the effect of peers’ characteristics. And third, which children do pupils interact with? I will present the results of two papers: my paper, What Makes a Test Score? and Hoxby and Weingarth’s paper Taking Race Out of the Equation: School Reassignment and the Structure of Peer Effects.
What Makes a Test Score? is joint work with Stephen Machin (University College London) and Francis Kramarz (CREST-INSEE). It estimates the sources of educational inequalities in the U.K. from to 11: do they come from schools, from segregation, or from differences in pupils’ backgrounds? We use the randomness of the demographics to estimate the effect of peers on achievement. Peer effects are significant but are smaller than what previous studies had shown.
Hoxby and Weingarth use a random school reassignment program in North Carolina. This provides a good source of random variation to estimate social interactions in the classrooms. Pupils benefit from their peers’ ability, provided it is not too different from their own ability.
Overall, confidence in the results relies on the use of credibly random variations. It brings clear public policy implications: peer effects matter, and a mix of segregation/integration is key to the improvement of education achievement in the classroom.
Francis Kramarz , Stephen J. Machin and Amine Ouazad, What Makes a Test Score? The Respective Contributions of Pupils, Schools, and Peers in Achievement in English Primary Education, INSEAD
Working Paper No. 2008/58/EPS.
CM Hoxby, G Weingarth, Taking Race Out of the Equation: School Reassignment and the Structure of Peer Effects, 2007, mimeo.