Misclassification and the Hidden Silent Rivalry

Abstract

The interaction of economic agents is one of the most important elements in economic analyses. While most empirical studies investigate peer effects on objective final achievements, peer effects on subjective outcomes are inherently difficult to identify and estimate because these variables are prone to measurement errors. In particular, peer effects on students' attitudes towards learning are believed to have a significant impact on their achievements, while we found the presence of misclassification errors in students' self-reported attitudes. We develop a binary choice model with misclassification and social interactions and use a recently developed technique of measurement error models to correct misreporting errors for estimating the peer effects on attitude. Our estimates suggest that a significant proportion of students overreport their attitudes towards learning and that peer effects are not only significant, but also much larger than estimates ignoring the misreporting errors. Our method may be generalized to the identification and estimation of peer effects with imperfect data information.

Identification and Inference of Network Formation Games with Misclassified Links

“…There are a few papers in the literature of social interactions that have examined the presence of measurement error in network data (Chandrasekhar and Lewis, 2014;Kline, 2015;Lewbel, Norris, Pendakur, and Qu, 2017;de Paula, Rasul, and Souza, 2018;Hu and Lin, 2018;Lewbel, Qu, and Tang, 2019). However, the results in these papers cannot be applied directly to our framework since they have a different object of interest.…”