The systematic measurement of socio-economic disadvantage has been significantly advanced in the past decade by the production of the Socio-Economic Indexes for Areas (SEIFA) by the Australian Bureau of Statistics. These measures are widely used by a number of government and nongovernment agencies in the allocation of funds on the basis of regional disadvantage, ranging from the Collection District to the State Divison levels. However, the methodology of SEIFA, while revealing national patterns of disadvantage, is based on two simplifying assumptions: (a) that the factorial structure of socio-economic disadvantage is invariate across regions and districts and (b) that it can be adequately captured by a single dimension, extracted by an unrotated Principal Components procedure. These assumptions are challenged in this paper, both theoretically and methodologically. A confirmatory factor analysis procedure, carried out on a combined sample of all Collection Districts in in the Northern Territory and Tasmania, revealed two strong orthogonal factors (representing economic and lifestyle/ identity features respectively). An introduced variable of regionality was also found to have a significant direct (non-spurious) contribution to a SEIFA-type index for this sample. These results suggest that the SEIFA procedure deserves further refinement in order to better represent the complexity and the context dependence of the underlying dimensions of disadvantage.
|Number of pages||26|
|Journal||Australian Journal of Social Research|
|Publication status||Published - Jan 1996|