Event

Large-scale data collection is costly, with the Australian Census costing $300 million in 2006, and $440 million in 2011.

Poor health outcomes are costly too, and not just in dollar terms. The burden of disease in Australia is led by tobacco in the majority of age groups, with Indigenous communities disproportionately figuring in this burden. Government policy to improve health outcomes in Indigenous communities is targeted at regions, and should be evaluated at a regional level. However data collections such as the National Health Survey and National Aboriginal and Torres Strait Islanders Social Survey are not detailed enough at the regional level to provide precise evaluations.

In this talk Alice will discuss a project that aims to

(1) estimate the prevalence of key health risk factors and outcomes with increased precision in sparsely sampled regions of Australia; and

(2) evaluate the change over time in key health risk factors and outcomes related to policies or interventions implemented at the regional (rather than a State or national) level. 

Candidate statistical methods that will be discussed comprise classical and Bayesian multilevel modelling. Similarities, differences and criticisms of these approaches will be discussed, along with relevant results from other data sources in Australia and other countries.

Alice Richardson is a biostatistician in the National Centre for Epidemiology & Population Health at the Australian National University. She studied at Victoria University of Wellington, New Zealand, then at the Australian National University, Canberra. Her PhD was on the statistical properties of robust methods of estimation for multilevel linear models.  She has twenty years of experience in teaching undergraduate statistics at the University of Canberra. During that time she also collaborated on quantitative research projects in every Faculty of the University. In 2016 she took up the position as biostatistician at ANU.

Her research interests are in linear models and robust statistics; statistical properties of data mining methods; applications of statistical methods to large data sets especially in population health and the biomedical sciences; and innovation in statistics education.