Precision of Compositional Data in a Stratified Two-Stage Cluster Sample: Comparison of the Swiss Earnings Structure Survey 2002 and 2004

• Activity Number: 415
• Type: Contributed
• Date/Time: Wednesday, August 9, 2006 : 10:30 AM to 12:20 PM
• Sponsor: Section on Survey Research Methods
• Author(s): Monique Graf*+
• Companies: Swiss Federal Statistical Office
• Keywords: complex survey ; compositional data ; linearization ; confidence domain ; coefficient of variation

Abstract

Precision of released figures is not only an important quality feature of official statistics, it is also essential for a good understanding of the data. In this paper we show a case study of how precision could be conveyed if the multivariate nature of data needs to be taken into account. In the official release of the Swiss Earnings Structure Survey, the total salary is broken down into several wage components. For surveys 2002 and 2004, we first investigate the incidence of components and then follow Aitchison's approach for the analysis of compositional data, which is based on logratios of components. Different multivariate analyses of the compositional data are performed and compared between the years, whereby the wage components are broken down by economic activity classes. Then we propose a number of ways to assess precision.