A design-based procedure for the analysis of experiments embedded in complex sample surveys

Experiments embedded in ongoing sample surveys are particularly appropriate to test effects of alternative surevey methodologies on estimates of finite population parameters.