|Page tools: Print Page Print All|
As part of the Census Data Enhancement project, the Australian Bureau of Statistics has conducted a quality study that simulates the formation of the Statistical Longitudinal Census Dataset (SLCD). This simulation has been carried out by linking 2006 Census and 2006 Census Dress Rehearsal data. The linking was carried out both with and without the use of name and address, the former acting as a benchmark for the latter. Linking without name and address is the method that is likely to be used for the planned linking of the 2006 and 2011 Censuses forming the first two waves of the SLCD. This paper describes a variety of methods that have been used to examine the quality of the data linked without name and address and extends those findings to predict the quality that can be expected when the first two waves of the SLCD are linked.
These documents will be presented in a new window.