This paper investigates the effect that covariate measurement error has on a conventional treatment effect analysis built on an unconfoundedness restriction that embodies conditional independence restrictions in which there is conditioning on error free covariates. The approach uses small...