Checking the Indexes Make Sense
Once the indexes are calculated, they are checked to ensure that they are measuring the desired concept and that the results make intuitive sense. This validation is vital to establish the credibility of the indexes and to identify any changes that need to be made prior to being released. The validation exercises carried out included:
- using a thematic mapping tool to view the spatial distribution of the scores
- consultation with ABS Regional Offices to validate the indexes against local knowledge
- investigation of the correlations between the four indexes
- identification of the most influential areas and variables in the index creation process, as identified by sensitivity analysis
- comparison of SEIFA 2011 rankings with 2006 rankings
- identification of the drivers of change from SEIFA 2006 to 2011 in terms of importance of particular variables
For a more in-depth discussion, please see Chapter 5 of the Technical Paper.
Other Issues to Consider
Topics not represented in the index
Users of the indexes should examine the variables contained in each index (see IRSD, IRSAD, IER, IEO) to ascertain whether they are appropriate for analysis. There are two aspects of advantage which the indexes do not represent well. Firstly, the indexes do not directly measure wealth, and secondly, there is very limited information available through the Census in regards to the infrastructure of an area. Both of these measures may provide more information about the level of advantage and disadvantage in an area. Other potential topics that could be associated with advantage and disadvantage but are not captured on the Census include health, crime, and the environment.
Area vs. Individual
The indexes reflect the socio-economic wellbeing of an area, rather than that of individuals. They were calculated at the SA1 level, and reflect SA1 characteristics. Because all people within a SA1 are not identical, the index scores for a SA1 do not directly reflect the relative advantage or disadvantage of an individual residing in that SA1. For example, it is possible for a relatively advantaged person to be a resident in a SA1 which may have a low score on some or all of the indexes. This needs to be taken into consideration when both using and validating the indexes. For further reading on finer level indexes and diversity within areas please see 'SEIFA: Getting a Handle on Individual Diversity Within Areas' (Cat. 1351.0.55.036)
Previous Page | Next Page