Exclusion Bias in the Estimation of Peer Effects

70 Pages Posted: 31 Aug 2016

See all articles by Bet Caeyers

Bet Caeyers

Institute for Fiscal Studies (IFS)

Marcel Fafchamps

Stanford University - Freeman Spogli Institute for International Studies

Date Written: August 2016

Abstract

We examine a largely unexplored source of downward bias in peer effect estimation, namely, exclusion bias. We derive formulas for the magnitude of the bias in tests of random peer assignment, and for the combined reflection and exclusion bias in peer effect estimation. We show how to consistently test random peer assignment and how to estimate and conduct consistent inference on peer effects without instruments. The method corrects for the presence of reflection and exclusion bias but imposes restrictions on correlated effects. It allows the joint estimation of endogenous and exogenous peer effects in situations where instruments are not available and cannot be constructed from the network matrix. We estimate endogenous and exogenous peer effects in two datasets where instrumental approaches fail because peer assignment is to mutually exclusive groups of identical size. We find significant evidence of positive peer effects in one, negative peer effects in the other. In both cases, ignoring exclusion bias would have led to incorrect inference. We also demonstrate how the same approach applies to autoregressive models.

Suggested Citation

Caeyers, Bet and Fafchamps, Marcel, Exclusion Bias in the Estimation of Peer Effects (August 2016). NBER Working Paper No. w22565, Available at SSRN: https://ssrn.com/abstract=2832569

Bet Caeyers (Contact Author)

Institute for Fiscal Studies (IFS) ( email )

7 Ridgmount Street
London, WC1E 7AE
United Kingdom

Marcel Fafchamps

Stanford University - Freeman Spogli Institute for International Studies ( email )

Stanford, CA 94305
United States

Here is the Coronavirus
related research on SSRN

Paper statistics

Downloads
6
Abstract Views
135
PlumX Metrics