Changes in the Selection of Dwellings in the Labour Force Survey of Argentina: A Simulation

• Activity Number: 247
• Type: Contributed
• Date/Time: Monday, August 5, 2013 : 2:00 PM to 3:50 PM
• Sponsor: Survey Research Methods Section
• Author(s): Augusto Hoszowski and Claudio Comari*+
• Companies: INDEC - Argentina and INDEC - Argentina
• Keywords: Sampling ; Variance Estimation ; Survey Methodololgy

Abstract

This paper summarizes the main considerations and results of the studies conducted in the Instituto Nacional de Estadística y Censos of Argentina, to evaluate the impact that the change in the methodology of selection of dwellings within each area (Secondary Sampling Unit) could have in the variances of certain estimators due to the passage from the method of selection of dwellings by systematic sampling to the method of selection of compact segments, in the household surveys. The results of a simulation exercise with microdata files from the 2010 Population Census are presented, comparing the different results obtained in relation to estimators for several population totals and their corresponding variances. It has been proven that going from systematic sampling to cluster sampling, during the last stage of selection, does not conduct to a significant increase in sampling errors, having therefore cluster sampling evident benefits in the organization of fieldwork.