Household surveys conducted by the Australian Bureau of Statistics (ABS) and other organisations are a key part of the statistical infrastructure of Australia. These surveys are affected by missing data at several levels: data items within people, people within households, or households as a whole. This project will create a framework and methods for handling missing data which reflect the multi-level nature and other complexities of household surveys. Existing methods do not incorporate these complexities. The new methods created will reduce biases and improve statistical analysis and inferences for household surveys. In addition, the project will provide high quality research training in applied statistics.
Household surveys conducted by the Australian Bureau of Statistics (ABS) and other organisations are a key part of the statistical infrastructure of Australia. These surveys are affected by missing data at several levels: data items within people, people within households, or households as a whole. This project will create a framework and methods for handling missing data which reflect the multi-level nature and other complexities of household surveys. Existing methods do not incorporate these complexities. The new methods created will reduce biases and improve statistical analysis and inferences for household surveys. In addition, the project will provide high quality research training in applied statistics.