Establishment Surveys With Population Survey–Generated Sampling Frames

Abstract

Abstract When stand‐alone sampling frames list all establishments and size measures, the Hansen–Hurwitz (HH) pps estimator is generally used to estimate the volume of transactions between establishments and households. The network sampling (NS) version of the HH estimator depends on a population‐survey‐generated establishment frame, which lists establishments that have transactions with households in a population sample survey and the number of transactions that establishments have with survey households. The NS estimator is a competitor of the HH estimator whenever stand‐alone frames of good quality are unavailable and, as this article indicates, the NS estimator may be competitive with the HH estimator when stand‐alone and population‐survey‐generated frames are flawless.