UC Davis Agricultural and Resource Economics

Marcel Fafchamps, Stanford University

Exclusion Bias in the Estimation of Peer Effects

Date and Location

Monday, April 11, 2016, 4:10 PM - 5:30 PM
ARE Library, 4101 Social Sciences and Humanities

Abstract

This paper formalizes a listed [Guryan et al., 2009] but unproven source of estimation bias in social interaction models. This bias is driven by the systematic exclusion of an individual from her peer group in the computation of average peer group outcomes. After deriving an exact formula for the magnitude of the bias in models using non-overlapping peer groups, we discuss its underlying parameters. We demonstrate that when the true peer e ffect is small or zero, the negative exclusion bias dominates the positive reflection bias yielding an overall negative bias on the peer e ffect estimate. We discuss the conditions under which the exclusion bias is aggravated when adding cluster fi xed eff ects. Simulation results confi rm all the theoretical predictions derived in this paper and illustrate how the bias a ffects inference and the interpretation of estimation results. To achieve consistent inference, we suggest correcting p-values using permutation methods. We provide a characterization of a generalized data generating process that can be used to consistently estimate all structural parameters in the model, both for models using peer groups and models using social network data. We also show the conditions under which two-stage least squares strategies do not suff er from exclusion bias. This may explain the counter-intuitive observation in the social interaction literature that OLS estimates of endogenous peer e ffects are often larger than their 2SLS counterparts.

Contact Us

2116 Social Sciences and Humanities
University of California, Davis
One Shields Avenue
Davis, CA 95616

Main Office: 530-752-1515

Student Advising Services: 530-754-9536