Measurement Errors in Introductory Econometric Courses

This article describes an instructive method for presenting the implications of measurement error in a regression analysis for the introductory econometrics course. As distinct from other methods, this method can be used to obtain estimates of the linear relationship between measured variables without requiring specific information on the relative magnitude of the measurement errors. However, the chosen method yields a set or range of estimates for a regression coefficient rather than a point estimate. This range, commonly referred to as bounds for a regression coefficient, indicates the uncertainty in a regression coefficient due to measurement error in one or more explanatory variables.