Get access

Identification of Peer Effects with Missing Peer Data: Evidence from Project STAR


  •  Corresponding author: Aaron Sojourner, Carlson School of Management, University of Minnesota, 321 19th Ave S, Minneapolis, MN 55455, USA. E-mail:

  • This article benefited from many helpful insights from Charles Manski, Christopher Taber and Greg Duncan. Thanks also to Charles Achilles, Jayne Boyd-Zaharias, Matthew Genztkow, Larry Hedges, Kyoo Il Kim, Colleen Manchester, Jason O'Connor, Steve Pischke, Christopher Rhoads, James Roberts, Elizabeth Tipton, Diane Whitmore Schazenbach, two anonymous referees and seminar participants at many venues for generous assistance at various stages. All errors are mine. This work was supported by a fellowship from the US Department of Education's Institute for Education Science.


This article studies peer effects on student achievement among first graders randomly assigned to classrooms in Tennessee's Project STAR. The analysis uses previously unexploited pre-assignment achievement measures available for 60% of students. Data are not missing at random, making identification challenging. This study develops and applies new ways to identify peer effects in the presence of missing data, which incorporate knowledge of how groups form. Estimates suggest sizeable positive effects of mean peer lagged achievement on average. Analysis of a common peer-effects estimator implies caution is warranted in interpreting many peer-effect estimates extant in the literature.