1352.0.55.054 - Research Paper: Strategies for Synthetic Estimation in ABS Business Surveys (Methodology Advisory Committee), Jun 2003  
ARCHIVED ISSUE Released at 11:30 AM (CANBERRA TIME) 20/05/2002  First Issue
   Page tools: Print Print Page Print all pages in this productPrint All
  • About this Release

About this Release

A number of external factors have contributed to a need to better understand the role of synthetic estimation methods in ABS business surveys:

  • ABS currently produces several statistical products for small subpopulations of the population of Australian businesses, in response to growing user demand for these products.
  • Sample sizes for direct collections by the ABS have decreased, for example the Manufacturing and Agricultural Censuses have become samples in the past five years. This is due to the need to reduce provider load and costs, and to an increased focus on national statistical priorities.
  • The quality and availability of taxation data have greatly improved. Business Income Tax (BIT) data has been in use for several years in the Economic Activity Survey (EAS). Business Activity Survey (BAS) data has been in limited use for several years for the purposes of sample design, and it has been used as an auxiliary variable for sample design or estimation in many surveys since August 2002.
  • An ABS small area estimates project is being conducted by MD. The project will produce a manual and a set of guidelines for applying small area estimation techniques in the ABS, by the end of 2003.

As a result, it is a good time to clarify the concepts underlying our current practices in synthetic or data substitution methods, and to suggest directions for future developments.