“…For matrix survey sampling with overlapping subsets of questions, Merkouris (2015) proposed an efficient estimation method based on the principle of best linear unbiased estimation, generates composite optimal regression estimators of population totals. Kamgar and Navvabpour (2017) proposed a strategy to split long questionnaires and estimate the population parameters based on data collected from subquestionnaires. They examined the four methods of small-area estimation, Horvitz-Thompson estimation, regression method, and multiple imputation, and concluded that of these four methods if the appropriate auxiliary variables are used from the registers or previous census, the small-area estimation method gives us a more accurate estimate.…”