Nonresponse in Rotating Panel Surveys: Analysis on Argentina's Labor Force Survey

• Activity Number: 339
• Type: Contributed
• Date/Time: Tuesday, August 5, 2014 : 10:30 AM to 12:20 PM
• Sponsor: Government Statistics Section
• Author(s): Augusto Hoszowski*+ and Claudio Comari
• Companies: INDEC and INDEC
• Keywords: sampling ; non response ; survey labor force

Abstract

The Encuesta Permanente de Hogares (EPH, for its acronym in spanish) is the official labor force survey of Argentina. The EPH's sample is divided in four different rotating panels, like in others surveys of its kind. The selected households stay for six quarters in the sample under the rotation schema known as 2-2-2. That is: two quarters into the sample, two quarters out and again two quarters into the sample, so each rotating panel is included for be interviewed in four times. In this paper we evaluate the effects of the use of rotating panels on the non response and the unemployment rate, using a data set of several consecutive quarters of the EPH. In order to do so, the information was organized according to the number of times each group of household was included to be interviewed. The observation of a higher non response rate in the panels included for first time than the observed in those included by several times was an unexpected result; as we show here the non response decrease for each panel along the successive participation into the sample