INTERCENSAL REVISION OF SLA POPULATION ESTIMATES
3.28 When SLA population estimates as at 30 June in a Census year are prepared, based on that year's Census, they can be compared with the preliminary estimates for the same date already produced by updating the previous Census (using the methods described above). Differences between these two sets of estimates are referred to as intercensal errors.
3.29 Preparing estimates for the same reference date but using different methods and sources can result in a break in the time-series of these estimates. To overcome the break in continuity between these two types of estimates all population estimates updated from the previous Census are recalculated. These estimates, referred to as rebased estimates, are derived for each SLA by apportioning the intercensal error evenly across the five years, while constraining the rounded SLA level estimates to sum to state and territory estimates. For example, rebased 2002 to 2005 estimates were derived by adding one-fifth of the 2006 intercensal error to the previous estimates of the 2002 population, two-fifths to the previous estimate of the 2003 population, and so on.
3.30 These intercensal total population estimates are broken down into age and sex components by interpolating ERP by cohort between the previous Census and current Census by SLA, age and sex (although for some particular age/sex cohorts within some SLAs the interpolation takes place based on the same age at the previous and current censuses, rather than a cohort-approach). All age/sex cells within each state/territory are forced to sum to the predetermined state/territory age/sex totals (see Appendix 5 - The iterative proportional fitting procedure).