Finite Population Mixed Models for Pretest-Posttest Designs with Response Errors

We consider a finite population mixed model that accommodates response errors and show how to obtain optimal estimators of the parameters in pretest-posttest context. illustrate method with estimation difference gain between two interventions simulation study compare empirical version proposed estimator (obtained by replacing variance components estimates) obtained via covariance analysis usually employed such settings. The results indicate many instances, has smaller mean squared error than from standard model.