作者: Eric A. Hanushek , John F. Kain , Jacob M. Markman , Steven G. Rivkin
DOI: 10.1002/JAE.741
关键词:
摘要: Empirical analysis of peer effects on student achievement has been open to question because the difficulties separating from other confounding influences. While most econometric attention directed at issues simultaneous determination interactions, we argue that omitted and mismeasured variables are likely be more important. We control for important determinants will confound estimates by removing school-by-grade fixed in addition observable family school characteristics. The also addresses reciprocal nature interactions interpretation based upon models using past as measure group quality. results indicate a positive effect growth. Moreover, students throughout test score distribution appear benefit higher achieving schoolmates. On hand, variance appears have no systematic effect. Copyright © 2003 John Wiley & Sons, Ltd.