3235.0 - Population by Age and Sex, Regions of Australia, 2016 Quality Declaration 
ARCHIVED ISSUE Released at 11:30 AM (CANBERRA TIME) 28/08/2017   
   Page tools: Print Print Page Print all pages in this productPrint All

Upcoming Change to Methodology for Sub-State Population Estimates

Sub-state population estimates for 30 June 2017 and onward will be prepared using a new, component-based methodology. This article discusses the reasons for the change, briefly describes the new methodology and outlines the ongoing release schedule for sub-state population estimates prepared using the component method.

The Component Method

Population changes as a result of natural increase (births minus deaths) and net migration (the movement of people to, from and within Australia). This is a fundamental demographic concept, known as the cohort component method, and is used by the ABS to estimate national, state and territory population change. The component method allows population change to be broken down into the components of natural increase and net migration, enabling a good understanding of population change for an area. More information on this method is provided in Population Estimates: Concepts, Sources and Methods (cat. no. 3228.0.55.001) and Australian Demographic Statistics (cat. no. 3101.0).

Reasons for Change

The component method has not previously been used to estimate population change at the sub-state (regional) level, due to the lack of comprehensive, high quality and timely estimates of migration at this level. Instead, sub-state population estimates have in the past been prepared using a regression-based approach, where population change is estimated based on changes in indicator data sources (e.g. dwelling approvals, and Medicare and Australian Electoral Commission enrolments). Regression modelling is a long-established method that has provided reasonable accuracy, however there are some weaknesses to this approach. For example, it is inconsistent with the method used to prepare national, state and territory estimates, and it doesn't provide an understanding of why population has changed.

The ABS has a long-established time series of births and deaths data at the sub-state level, and has invested significant effort in recent years to develop an effective and timely method of estimating regional internal migration. This has allowed a model to estimate the remaining component - regional overseas migration - to be developed. As such, the ABS can now implement the component method for sub-state population estimates. This change will bring consistency with the method used to prepare national, state and territory population estimates and provide a much richer picture of regional population change than is currently available. A summary of the data sources for each of the components at the sub-state level, by age and sex, is provided below.

Sub-state component data sources

    Births and Deaths
    Births and deaths data are provided to the ABS by state and territory Registrars of Births, Deaths and Marriages. The registration of births and deaths is compulsory in Australia under relevant state/territory legislation. This data is provided by age and sex and is geocoded to the required levels of the Australian Statistical Geography Standard (ASGS) for sub-state ERP production - namely Statistical Areas Level 2 (SA2s) and Local Government Areas (LGAs).

    Internal Migration
    Regional internal migration estimates (RIME) are prepared using Medicare change-of-address data by age and sex, provided by the Australian Government Department of Human Services. Expansion factors are applied to account for moves by age and sex, state/territory and type of move (arrival or departure) known to be missing from the Medicare data (e.g. young adult males are less likely to update their details with Medicare). This data is supplemented with aggregated data on Australian Defence Force personnel movements, supplied by the Australian Government Department of Defence.

    In the past, this Medicare data was converted from the postcode level to SA2 and LGA regions. For further information on this method, see Discussion Paper: Assessment of Methods for Developing Historical Estimates of Regional Internal Migration (cat. no. 3405.0.55.001). From 2017 onwards, the ABS will instead make use of Medicare data coded directly to the ASGS, significantly simplifying the method outlined in this paper.

    Overseas Migration
    The ABS has developed a model to estimate total overseas migration into and out of the regions, utilising what we know about population change from previous Censuses, and from the natural increase and RIME components. The main data source used to model these regional overseas migration estimates (ROME) is the Census. In particular, information on the number of overseas arrivals to regions, the proportion of the total population of a region born overseas and Socio Economic Indexes For Areas (SEIFA) scores for regions. These estimates can then be broken down into age and sex, utilising information from the Census and other data sources at the age and sex level.

Preparation and Release Schedule

The ABS will continue to release annual sub-state population estimates as at 30 June for regions of the ASGS including SA2s and LGAs. The method for Census year estimates will not change. That is, estimates will be based on Census counts of usual residence (excluding short-term overseas visitors in Australia), with an allowance for Census net undercount and Australian residents temporarily overseas at the time of the Census.

Estimates in non-Census years will be prepared by adding natural increase (births minus deaths) and net internal and overseas migration by age and sex to the base population. Each of these components, which are derived from the data sources discussed above, will be constrained to the corresponding state and territory estimates by age and sex released in Australian Demographic Statistics (cat. no. 3101.0) to ensure consistency between state and territory and sub-state population estimates.

The release of estimates prepared using the component method will not change, with totals released around 9 months after the reference date and age-sex breakdowns available roughly 13 months after the reference date. The first sub-state total population estimates prepared using this new method will be for 30 June 2017 and will be released in Regional Population Growth (cat. no. 3218.0) on 28 March 2018. In addition to the population estimates, each of the components of population change for SA2s and LGAs for 2016-17 will be included, allowing users to understand the main drivers behind population change for sub-state areas. Age-sex breakdowns of these estimates will then be released in late August 2018 in Population by Age and Sex, Regions of Australia (cat. no. 3235.0).

For further information on this new methodology, please contact the Regional Population Unit at regional.population@abs.gov.au.