ERP BY AGE AND SEX
4.17 Once the total population estimates are prepared for CDs, they are later broken down into age and sex components. The method to calculate ERP by age and sex for customised geography converts Census usual residence CD counts by age and sex into CD ERP by age and sex, i.e. combining adjustments for net Census undercount, RTOs and backdating components from Census Night to 30 June. In post-censal years, forcing (using iterative proportional fitting, see Appendix 5 - The iterative proportional fitting procedure) to SLA ERP and CD totals with growth applied implicitly adjusts the CD age/sex population for post-censal births, deaths, migration and ageing.
4.18 Calculate CD of usual residence population counts by age and sex as at Census Night (from Census data), and adjust for any known issues with Census data. This forms the body of the matrix for iterative proportional fitting.
4.19 Adjust the CD ERP for post-censal years to reflect population growth from indicator data (in 2007 electoral roll data was used). This forms the column marginal in the matrix for iterative proportional fitting.
4.20 Extract SLA ERP by age and sex from those published in Population by Age and Sex, Regions of Australia (cat. no. 3235.0). This forms the row marginal in the matrix for iterative proportional fitting.
4.21 Using iterative proportional fitting, force the CD age/sex distribution to sum to the CD totals and SLA age/sex ERPs.
4.22 Aggregate CDs to the required geography.