Premium
Measurement of Peer Effects
Author(s) -
Ryan Chris
Publication year - 2017
Publication title -
australian economic review
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.308
H-Index - 29
eISSN - 1467-8462
pISSN - 0004-9018
DOI - 10.1111/1467-8462.12213
Subject(s) - affect (linguistics) , sort , sorting , point (geometry) , face (sociological concept) , reflection (computer programming) , peer group , measure (data warehouse) , psychology , computer science , social psychology , sociology , mathematics , social science , data mining , information retrieval , geometry , communication , programming language
Papers that aim to measure peer effects on the educational outcomes of an individual face at least two problems in estimating them convincingly. The first is a sorting problem: in education, individuals who are already alike tend to sort together into similar schools or institutions. The second is the reflection problem: if my peers affect me, do I not also affect them? Solutions to these problems are not straightforward and the methods often used in the literature are contentious, as this review aims to make clear. External manipulation of peer groups, via quasi‐ or full randomisation, provides a necessary starting point.