Premium
Causal Inference with Differential Measurement Error: Nonparametric Identification and Sensitivity Analysis
Author(s) -
Imai Kosuke,
Yamamoto Teppei
Publication year - 2010
Publication title -
american journal of political science
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 6.347
H-Index - 170
eISSN - 1540-5907
pISSN - 0092-5853
DOI - 10.1111/j.1540-5907.2010.00446.x
Subject(s) - observational error , causal inference , nonparametric statistics , identification (biology) , inference , robustness (evolution) , econometrics , differential (mechanical device) , errors in variables models , computer science , sensitivity (control systems) , statistics , mathematics , artificial intelligence , botany , electronic engineering , engineering , biology , aerospace engineering , biochemistry , chemistry , gene
Political scientists have long been concerned about the validity of survey measurements. Although many have studied classical measurement error in linear regression models where the error is assumed to arise completely at random, in a number of situations the error may be correlated with the outcome. We analyze the impact of differential measurement error on causal estimation. The proposed nonparametric identification analysis avoids arbitrary modeling decisions and formally characterizes the roles of different assumptions. We show the serious consequences of differential misclassification and offer a new sensitivity analysis that allows researchers to evaluate the robustness of their conclusions. Our methods are motivated by a field experiment on democratic deliberations, in which one set of estimates potentially suffers from differential misclassification. We show that an analysis ignoring differential measurement error may considerably overestimate the causal effects. This finding contrasts with the case of classical measurement error, which always yields attenuation bias.