|Page tools: Print Page Print All|
FEATURE ARTICLE: RECASTING 20 YEARS OF AUSTRALIA'S SUB-STATE POPULATION ESTIMATES
WHY RECASTING WAS NECESSARY
After each Census, the ABS rebases ERP using the latest Census count. As part of this process, ERP is adjusted to take into account any intercensal error by distributing the error evenly across the five year period since the previous Census.
Preliminary rebasing of ERP after the 2011 Census followed this standard methodology, with sub-state population published in July-August 2012. However, the standard rebasing treatment could not credibly account for the large intercensal error between 2006 and 2011, and the resulting ERP series showed implausible growth over this period for many regions. The large intercensal error identified was predominantly due to a change in methodology in the 2011 Census Post Enumeration Survey (PES).
The PES is conducted after each Census to assess coverage of Census counts, as represented by the key measure of net undercount. The 2011 PES utilised the new methodology of Automated Data Linking (ADL), which resulted in better linking and matching of PES and Census records, and a better measure of net undercount.
A statistical impact study conducted to determine the impact of using the new ADL method found that the 2011 net undercount was approximately 40% lower than it would have been if the previous methods had been used. The ABS estimated that the previous net undercount would have been substantially lower for all previous post enumeration surveys had this methodology been available, and the large intercensal error in 2011 is therefore likely to have been accumulated over a period greater than the usual five years. It was decided that this large statistical impact (which accounts for around 84% of the 2006-2011 national intercensal error) should be incorporated into ERP by distributing it over a period longer than the usual five years.
THE DECISION MAKING PROCESS
The ABS intention to recast ERP data over a period longer than the usual five year rebasing period was first proposed in the 27 September 2012 release of Australian Demographic Statistics, along with an open invitation for comment. After an extensive consultation process, the ABS made the decision to recast ERP over a period of 20 years. This was announced in the 18 December 2012 release of the same publication. Output plans for the recast sub-state ERP series were announced in the 30 April 2013 release of Regional Population Growth.
The 20 year period (reflecting four intercensal periods) was decided on as the period of time that would result in an estimate of population growth that reflected the growth observed in the historical data for population components (that is, births, deaths and migration), which are the best data source for measuring population change over time at the national and state/territory level, and where the recast estimates could be confidently broken down below state/territory level.
METHODOLOGY USED IN RECASTING
The processes and methods used to recast the data were developed and quality assured by a team of demographic and methodological specialists within the ABS. These methods were guided by a series of principles that were developed during the consultation process, which established that:
2. The use of ADL in the PES has been a major improvement in how the ABS measures Census coverage, and the 2011 net undercount should be used to inform historical understanding of Census coverage.
3. Population growth for the 2006-2011 period should, as closely as possible, reflect the growth in the population components (i.e. births, deaths and migration) or other indicator data for all spatial levels (i.e. national, state, and sub-state).
4. Any assumptions should be based upon the best available data.
5. Any revision to the historical ERP series should maintain the demographically plausible relationships between the fundamental building blocks of population series (e.g. age-sex profiles).
6. Where revised data exist for population components data, they should be used regardless of whether they were available at the time of previous rebasing processes.
Census point adjustments
The usual rebasing treatment of intercensal error distributes the error evenly over the five year period, as there is no further data to inform upon the distribution of the error within the period. Rather than distributing the impact of ADL back evenly over the 20 year period, the recasting process differentially adjusted each of the four intercensal periods, based on all information available for that period. This resulted in a greater impact on the data around the 2006 Census point, gradually decreasing to a minimal impact on the 1991-1996 data.
The recasting process involved calculating revised estimates at each Census rebasing point (ERP at 30 June of each Census year 2006, 2001 and 1996), with these points then used as the base population for the rebasing of annual intercensal estimates between these recast base points (and the 1991 Census-based ERP). This recasting process therefore only involved change in the ERP series, and effectively adjusted the previous estimates of undercount from these Censuses. No adjustment was made to the actual Census counts.
Adjustment calculations followed a top-down approach; the first step was to calculate the total adjustment at a national level for each Census point. This was then apportioned to the states and territories, followed by part of state (Greater Capital City Statistical Area and Rest of state/territory), then the lower levels of geography. Age and sex profiles were calculated and applied following the same top-down sequence.
National and state/territory adjustments
The national adjustment in 2006 was taken directly from the ADL statistical impact study results. For 2001, the reduction was based on the 2006 reduction and an offsetting impact of a change in PES methodology made between the 2001 and 2006 surveys. The 1996 national adjustment was derived to minimise the change between the previously published estimates and recast estimates between 1991 and 2001, and which would maintain growth rates over the 1991-1996 and 1996-2001 periods as closely as possible.
The apportionment of the national adjustment to the state and territory level was calculated based on a combination of intercensal error and the 2006 PES adjustment. To derive the state and territory level adjustments for the 2001 Census point, the 2006 state and territory split was offset by the state and territory specific impact of the change in PES methodology made between the 2001 and 2006 surveys. The 2001 state and territory split was then multiplied by the ratio between the 1996 and 2001 magnitude adjustments to derive the 1996 state and territory split.
For more information about the national and state/territory adjustments refer to the 20 June 2013 release of Australian Demographic Statistics, Feature Article 2: Recasting 20 years of ERP.
Adjustments to the annual (30 June) series of ERP at lower levels of geography involved apportioning the state/territory level adjustments to each part of state (i.e. Capital city and Rest of state regions), and then further apportioning the adjustments to the component regions within them.
Recast ERP for 30 June 2006, 2001 and 1996 was prepared for each part of state based on a methodology that drew upon a combination of component growth for each intercensal period, and changes in Census counts supplemented by estimates of residents temporarily overseas on each Census night.
This method provided a reliable and consistent measure of population change by part of state over this 20 year period and enabled the ABS to derive a quality set of adjustments. In particular these adjustments resulted in a much greater degree of internal consistency in the time series of sub-state ERP than was previously possible. Table 1 shows the recasting adjustments made to each part of state for Census years 2006, 2001 and 1996, i.e. compared with the original, previously published version of these estimates as released in July 2012.
All data was prepared on consistent part of state boundaries, based on the 2011 edition of the Australian Statistical Geography Standard (ASGS).
The new estimates for each part of state for 2006, 2001 and 1996 were then disaggregated into lower levels of geography (for example, SA2s and LGAs) based on the previously published breakdown of ERP on these geographies. The previously published estimates were based on usual residence counts from the 2006, 2001 and 1996 Censuses, all converted to 2011 boundaries.
Recasting annual estimates
With the adjusted population levels set for each Census point, the annual intercensal estimates were then recast progressively at the state and territory level from the 1991-1996 to 2006-2011 periods according to the standard rebasing method. This method continued to assume that the intercensal error should be apportioned equally over the five-year intercensal period.
At the part of state level, component-based estimates of annual population change between Census years were applied to obtain 30 June population estimates for the intercensal years. The following graph illustrates that at the combined part of state level the recast series is noticeably less variable than the previously published series. This is also a positive feature of the recast series for most regions in Australia.
New estimates for each part of state at 30 June 1992 to 2006 were then disaggregated into lower levels of geography based on the previously published disaggregations of these geographies. For instance, where Walkerville had 0.603% of the population of Greater Adelaide in 2006, after recasting it still had 0.603% of the population.
For some areas, further adjustments were made to resolve anomalies in previously published population change for these regions. These one-off adjustments are not considered as part of the recasting process, but rather as refinements to historical estimates converted from previous geographies to the new ASGS regions.
Where possible, population estimates for groups of regions based on different geographies (for example, SA2 and LGA regions) were aligned.
New estimates for each region were in turn broken down into age and sex based on the previously published age and sex structure of each region.
IMPACT ON THE DATA AND ITS USE
When comparing the original ERP series with the recast series, the distribution of population across regions at each previous point in time has not changed substantially. This is because Census counts (which have not changed) still account for the majority of data used to prepare regional population estimates for Census years, with indicator data still used to estimate intercensal change by year.
However, due to the varying degrees of population change across regions, the relatively similar adjustments made for each region has resulted in varying degrees of changes to growth rates across the regions for the same time period. This is especially true for areas previously considered as having low or negative change. Regions previously regarded as high-growth are generally still regarded as high-growing areas. Examples (for three areas within Rest of Victoria) are presented in table 2.
To facilitate users assessing changes between the previously published and recast series, the ABS has ensured the consistent formatting of data cubes released with preliminary rebasing in July-August 2012 and final rebasing on 30 August 2013. To determine the change resulting from recasting, the ABS advises subtracting the new estimates from the previously published estimates for regions of interest.
For information on the revisions to other data resulting from this recasting process refer to the 20 June 2013 release of Australian Demographic Statistics, Feature Article 2: Recasting 20 years of ERP.
It is expected that in the future, ERP will continue to be rebased on a five yearly basis after each Census. In contrast, recasting should be seen as an exceptional event made necessary by the significant methodological improvement in 2011; it is not anticipated that the recasting process will be repeated in relation to ERP in the foreseeable future. In addition, the experience gained from the recasting process and its impact on ERP will be used to inform plans for the 2016 rebasing and beyond.
For more information, contact Andrew Howe on (08) 8237 7370, or email email@example.com.
These documents will be presented in a new window.