Efficiencies in Variance Estimation for Complex Survey Data

Abstract

Abstract Data from complex survey designs require special consideration, with regard to variance estimation and analysis, because of design components that include unequal selection probabilities, stratification, and clustering. Using data from the National Medical Care Expenditure Survey, which is characterized by a complex survey design, four variance estimation programs are compared: SESUDAAN/RATIOEST, SUPERCARP, PSALMS, and HESBRR. The comparisons concentrate on program capabilities, computational efficiency, and user facility.