UPCOMING REVISIONS TO POPULATION ESTIMATES


INTRODUCTION

The ABS recently rebased population estimates based on results from the 2011 Census of Population and Housing. This involved creating a new 2011 base and using this to revise the estimated resident population (ERP) between the new Census and the 2006 Census. The new 2011 base was calculated using usual resident counts from the 2011 Census, together with an estimate of net undercount (as measured by the Census Post Enumeration Survey, or PES), residents temporarily overseas on Census night, and an adjustment to backdate from Census night to 30 June.

In 2011, the ABS introduced a new method of processing the PES which resulted in an improved, and lower, measure of net undercount than would have been derived had the same methodology been used in processing the 2006 PES. In other words, the 2011 PES found that more people were actually counted in the Census than would have been suggested by the previous PES methodology.

As a result of this, the net undercount adjustment applied in the rebasing of 2011 population estimates was considerably lower than the undercount adjustment applied in rebasing 2006 population estimates. This contributed to an intercensal error for 2006-11 of 294,000, which was significantly higher than the intercensal errors that have been measured historically.

The conventional demographic treatment for intercensal error is to spread the total amount evenly through the ERP series for the previous five years. However, due to the relatively high intercensal error for the 2006-11 period caused by the improved PES processing, the ABS proposed adjusting historical ERP over a longer period to reflect the different undercount adjustments that would have been used had the improved PES methodology been available in earlier years. After an extensive consultation process, the ABS determined that backcasting 20 years of population estimates back to 1991 will be undertaken, and incorporated into the final 2011 rebasing process later in 2013. For more information, see:

• The 2006-11 Intercensal Period and Revisions to Historical ERP
• Basic Guide to Historical Population Estimates in 2013

Accordingly, the population estimates released in this and previous issues of this product will be updated during final rebasing. A summary of the different revision processes is provided below.


REVISIONS TO HISTORICAL ERP

The 2006-11 intercensal error will be distributed over a 20 year period, resulting in a revision to the historical ERP series at the national, state and territory, and sub-state levels. This will involve effectively revising the PES undercount adjustment of the ERP rebasing process for the 1996, 2001 and 2006 bases. The revisions will have the largest impact on 2006 ERP and the smallest impact on 1996 ERP.

At the national and state/territory level, the intercensal error will be spread amongst quarterly population estimates from March 2011 back to September 1991.

The annual series of sub-state SA2-based estimates (as at 30 June) will be revised from 2010 back to 1992. The availability of revised estimates for the various sub-state geographies will differ depending upon the type of estimate (total or age-sex) and past availability. Further information on this is contained later in this article.


ADDITIONAL 2011 REVISIONS

During rebasing, estimates of net undercount are applied at the part of state level. To further apportion this undercount to sub-part of state regions, the age and sex structure of each region, together with the proportion of the region's population that is Aboriginal or Torres Strait Islander, is taken into account.

As the improvements to the PES have resulted in a much lower net undercount, the methodology that was used to apportion the undercount amongst sub-state regions has also been re-evaluated, to determine whether it is continuing to provide the most robust estimates at lower levels of geography, ahead of final rebased estimates being released on 30 August 2013. The results of this analysis showed that too much undercount appeared to be allocated to areas with a large proportion of Aboriginal and Torres Strait Islander peoples, above what is considered to be demographically plausible, while less was allocated to those areas with a small proportion.

While undercount constitutes only a relatively small proportion of the population estimates, the ABS has determined that this methodology will be refined for final rebasing, and some minor revisions will be incorporated into final rebased sub-state 2011 estimates. Sub-state estimates of the Aboriginal and Torres Strait Islander population will also be produced after this process is complete, and these 2011 estimates will also be released on 30 August 2013.

Once 2011 population estimates are finalised, 2012 sub-state estimates will also be revised and released on 30 August 2013. The 2011-12 growth rates released in this issue of Regional Population Growth, Australia, 2011-12 (cat. no. 3218.0) will generally be applied to the new 2011 base to give updated 2012 estimates.


WHEN WILL REVISED DATA BE RELEASED?

National and State/Territory

Revised September 1991 to March 2011 population estimates for Australia and for states and territories will be released in Australian Demographic Statistics, Dec 2012 (cat. no. 3101.0) on 20 June 2013. Total ERP (by sex) will be released for each quarter, while age-sex breakdowns will be released for 30 June of each year.


Sub-state

The following revised sub-state ERP totals will be released on the ABS website on 30 August 2013 in Regional Population Growth, Australia, 2012 (cat. no. 3218.0):

	1991-2000	2001-2010	2011	2012
SA2 & SA2-based(a)	P	P	P	P
LGA(b)	O	P	P	P
SA1-based(c)	O	P	P	P
SLA & SLA-based(d)	O	P	P	O
(a) Statistical Area Level 2-4 (SA2-SA4), Greater Capital City Statistical Area and Significant Urban Area, based on ASGS2011 boundaries. (b) Local Government Area (LGA) based on ASGS2012 boundaries. (c) Remoteness Area and Commonwealth and State Electoral Divisions, based on ASGS2011 boundaries. (d) Statistical Local Area (SLA), Local Government Area, Statistical Subdivision, Statistical Division and Statistical District, based on ASGC2011 boundaries.

On the same day, the following revised total ERP will become available upon request as a charged consultancy:

	1991-2000	2001-2010	2011	2012
SA1 & SA1-based	O	P	P	P
CD & CD-based	O	P	P	O
(a) Statistical Area Level 1 (SA1), Postal Area, State Suburb, Urban Centre/Locality and Section of State, based on ASGS2011 boundaries. (b) Census Collection District (CD), Postal Area, State Suburb, Remoteness Area, Commonwealth and State Electoral Divisions, Urban Centre/Locality and Section of State, based on ASGC2006 boundaries.

The following revised sub-state ERP by age and sex will also be released on 30 August 2013 in Population by Age and Sex, Regions of Australia, 2012 (cat. no. 3235.0):

	1991-2000	2001-2010	2011	2012
SA2 & SA2-based(a)	O	P	P	P
LGA(b)	O	P	P	P
SLA & SLA-based(c)	O	P	P	O
(a) Statistical Area Level 2-4 (SA2-SA4) and Greater Capital City Statistical Area, based on ASGS2011 boundaries. (b) Local Government Area (LGA) based on ASGS2012 boundaries. (c) Statistical Local Area (SLA), Local Government Area, Statistical Subdivision and Statistical Division, based on ASGC2011 boundaries.

The following revised age-sex ERP will become available as a charged consultancy upon request at the same time:

	1991-2000	2001-2010	2011	2012
SA1 & SA1-based(a)	O	O	P	P
CD & CD-based(b)	O	P	P	O
(a) Statistical Area Level 1 (SA1), Postal Area, State Suburb, Remoteness Area, Commonwealth and State Electoral Divisions, Urban Centre/Locality and Section of State, based on ASGS2011 boundaries. (b) Census Collection District (CD), Postal Area, State Suburb, Remoteness Area, Commonwealth and State Electoral Divisions, Urban Centre/Locality and Section of State, based on ASGC2006 boundaries.

For data requests, email client.services@abs.gov.au. For other queries related to sub-state ERP, email regional.population@abs.gov.au.