Estimates of Aboriginal and Torres Strait Islander Australians methodology

1 This publication contains final estimates of the Aboriginal and Torres Strait Islander, non-Indigenous and total populations of Australia at 30 June 2016, based on results of the 2016 Census of Population and Housing.

2 Estimates are disaggregated by age and sex for states/territories, Remoteness Areas and Indigenous Regions. Only total estimates for Statistical Areas Level 2 are available.

3 Estimated resident population (ERP) is the official measure of the population of Australia, based on the concept of usual residence within Australia. Usual residence is that place where each person has lived or intends to live for six months or more from the reference date for data collection.

4 The estimates of the Aboriginal and Torres Strait Islander population presented in this publication are based on 2016 Census of Population and Housing counts of Aboriginal and Torres Strait Islander Australians, adjusted for net undercount as measured by the Post Enumeration Survey. The extent of undercoverage of Aboriginal and Torres Strait Islander Australians in the 2016 Census and the relatively small sample size of the Post Enumeration Survey to adjust for that undercoverage means the estimates should be interpreted with a degree of caution. For more information on data quality see the relevant section of this Explanatory Note and Technical Note: Estimated Resident Aboriginal and Torres Strait Islander Population - Method of Calculation.

5 The Aboriginal and Torres Strait Islander population of Australia comprises people who are of Aboriginal origin, Torres Strait Islander origin or both Aboriginal and Torres Strait Islander origin. The Commonwealth definition of an Aboriginal and/or Torres Strait Islander person is:

• a person of Aboriginal and/or Torres Strait Islander descent who;
• identifies as being of Aboriginal and/or Torres Strait Islander origin and who is;
• accepted as such by the community with which the person associates.
   

6 The 2016 Census of Population and Housing (Household Form) asked the following question of each person

7 The final Aboriginal and Torres Strait Islander population estimate is based on the 2016 Census of Population and Housing and is calculated using a similar process to that used for the total Australian population.

• We calculate the final rebased Australian population estimate at 30 June in a Census year by:
• Adjusting Census counts of Australian usual residents to account for people missed, counted more than once, or counted in error in the Census (based on PES results)
• Demographically adjusting Census counts or net undercount estimates based on coherence with other data sources and to remove implausible demographic features
• Adding usual residents temporarily overseas on Census night
• Adjusting for births, deaths and migration from 30 June to Census night.

8 For further information, see Technical Note: Estimated Resident Aboriginal and Torres Strait Islander Population - Method of Calculation.

9 The ABS aims to produce high quality data from the Census. To achieve this, extensive effort is put into Census form design, collection procedures, and processing procedures. There are four principle sources of error in Census data: partial response, processing error, respondent error and undercount.

10 Partial response: When completing their Census form, some people do not answer all the questions which apply to them. While questions of a sensitive nature are generally excluded from the Census, all topics have a level of non-response. However, this level can be measured and is generally low. In those instances where a householder fails to answer a question, a not stated code is allocated during processing, with the exception of non-response to age, sex, marital status and place of usual residence. These variables are needed for population estimates, so they are imputed using other information on the Census form, as well as aggregate data from the previous Census.

11 Processing error: The processing of information from Census forms is mostly automated, using scanning, Intelligent Character Recognition and other automatic processes. Quality assurance procedures are used during Census processing to ensure processing errors are kept at an acceptable level. Sample checking is undertaken during coding operations, and corrections are made where necessary.

12 Respondent error: The Census form may be completed by one household member on behalf of others. Incorrect answers can be introduced to the Census form if the respondent does not understand the question or does not know the correct information about other household members. Many of these errors remain in the final data.

13 Undercount: The goal of the Census is to obtain a complete measure of the number and characteristics of people in Australia on Census night and their dwellings, but it is inevitable that a small number of people will be missed, some will be counted more than once and some will not be identified correctly. In Australia, more people are missed from the Census than are counted more than once. The net effect when both factors are taken into account is an undercount.

14 Each of these sources of error are particularly relevant to, and have the potential to significantly impact on, the Census counts of Aboriginal and Torres Strait Islander Australians. This is despite the implementation of the Discrete Community and Remote Areas Strategy, and specific urban strategies that included procedures to improve coverage, accuracy and quality of Aboriginal and Torres Strait Islander counts. For further information see Census of Population and Housing - Counts of Aboriginal and Torres Strait Islander Australians (cat. no. 2075.0).

15 Further information on Census data quality is available in Census Dictionary (cat. no. 2901.0) and 2016 Census Data Quality Working Papers, available on the ABS web site www.abs.gov.au/census.

16 The ABS conducts the PES shortly after the Census to determine how many people were missed in the Census and how many were counted more than once. The design of the survey is such that estimates of net undercount are suitable for augmenting Census counts for the purpose of deriving population estimates for Australia and the states and territories. For 2016, the survey had a sample size of around 42,000 households across Australia.

17 As estimates of undercount are based on a sample survey they are subject to sampling error. Since only a sample of dwellings is included in the PES, estimates derived from the survey may differ from figures which would have been obtained if all dwellings had been included in the survey. One measure of the likely difference is given by the standard error (SE) which indicates the extent to which an estimate might have varied by chance because only a sample was included. The relative standard error (RSE) is the standard error expressed as a percentage of the estimate to which it refers.

18 For Australia, the direct estimate of the Aboriginal and Torres Strait Islander population obtained from the PES was 786,689 persons, with a standard error of 19,776 (and a relative standard error (RSE) of 2.1%). For estimating the Aboriginal and Torres Strait Islander population of states and territories, undercount estimates were adjusted to improve the reliability of estimates. For more information on the method applied, see Technical Note: Estimated Resident Aboriginal and Torres Strait Islander Population - Method of Calculation. Unadjusted undercount rates are presented below but should be used for illustrative purposes only. The adjusted estimates which were actually used in the calculation of final Aboriginal and Torres Strait Islander population estimates and which should be used, can be found in Technical Note: Estimated Resident Aboriginal and Torres Strait Islander Population - Method of Calculation.

19 The PES sample is insufficient to produce estimates of net undercount by Indigenous status at the sub-state/territory level. Undercount was therefore apportioned to Statistical Areas Level 2 based on age, sex, Indigenous status and state/territory. For further information on this process, please refer to Regional Population Growth, Australia, 2016-17 (cat. no. 3218.0).

20 It is important to note that at the sub-state/territory level, differences between Census counts and estimates of the Aboriginal and Torres Strait Islander population should not be interpreted as a definitive measure of the actual level of undercount; rather, these differences are a by-product of the assumptions that contribute to the estimation process, and the differences should be considered indicative, based on the best available (though limited) information.

21 For further information see Technical Note: Estimated Resident Aboriginal and Torres Strait Islander Population - Method of Calculation.

22 For further information on the Post Enumeration Survey see Census of Population and Housing: Details of Overcount and Undercount, Australia (cat. no. 2940.0).

23 This publication contains data coded to a number of statistical geographic structures within the statistical geography classification called the Australian Statistical Geography Standard (ASGS). The structures used are: the Main Structure, the Remoteness Structure and the Indigenous Structure. Information on the ASGS is available on the ABS website under 'Statistical Geography'. Alternatively, see Volume 1 - Main Structure and Greater Capital City Statistical Areas (cat. no. 1270.0.55.001).

24 The ABS Structures are a hierarchy of regions developed for the release of particular ABS statistics. Their components are described below:

• Mesh Blocks (MBs): are the smallest geographical region in the ASGS. The 2016 ASGS contains 358,122 MBs covering the whole of Australia. They broadly identify land use such as: residential, commercial, agriculture and parks etc. As Mesh Blocks are very small they can be combined together to accurately approximate a large range of other statistical regions.
• Statistical Areas Level 1 (SA1s): have been designed as the smallest unit for the release of Census data. SA1s generally have a population of 200 to 800 persons, and an average population of about 400 persons. They are built from whole Mesh Blocks, and there are 57,523 SA1s covering the whole of Australia.
• Statistical Areas Level 2 (SA2s): are a general-purpose medium sized area built from whole SA1s. Their aim is to represent a community that interacts together socially and economically. SA2s generally have a population range of 3,000 to 25,000 persons, and have an average population of about 10,000 persons. The SA2 is the lowest level of the ASGS structure for which Estimated Resident Population (ERP), Health and Vitals and other non-Census ABS data are generally available. There are 2,310 SA2s covering the whole of Australia.
• Statistical Areas Level 3 (SA3s): provide a standardised regional breakup of Australia. The aim of SA3s is to create a standard framework for the analysis of ABS data at the regional level through clustering groups of SA2s that have similar regional characteristics. SA3s are built from whole SA2s and in general have populations between 30,000 to 130,000. They are often the functional areas of regional cities and large urban transport and service hubs. There are 358 SA3 regions covering the whole of Australia.
• Statistical Areas Level 4 (SA4s): are the largest sub-State regions in the Main Structure. They are designed for the output of Labour Force Survey data and reflect labour markets within each state and territory. SA4s are built from whole SA3s and cover the whole of Australia. There are 107 SA4 regions covering the whole of Australia.
   

25 For the purposes of this publication, Jervis Bay Territory, Christmas Island, Cocos (Keeling) Islands and Norfolk Island are collectively referred to as 'Other Territories' and are included in the totals for Australia. For further information see Australian Statistical Geography Standard (ASGS): Volume 1 - Main Structure and Greater Capital City Statistical Areas (cat. no. 1270.0.55.001).

26 The Remoteness Structure divides Australia into broad geographic regions that share common characteristics of remoteness in relation to access to services. There are six classes of RA in the Remoteness Structure: Major Cities of Australia, Inner Regional Australia, Outer Regional Australia, Remote Australia, Very Remote Australia and Migratory.

27 Within each state/territory, each RA represents an aggregation of non-contiguous geographical areas which share common characteristics of remoteness (constructed from SA1s). While statistical data classed to this structure may be available by state/territory, characteristics of remoteness are determined in the context of Australia as a whole. Therefore, not all RAs are represented in each state/territory.

28 For further information see Australian Statistical Geography Standard (ASGS): Volume 5 - Remoteness Structure (cat. no. 1270.0.55.005).

29 Data are also presented according to the Indigenous Structure of the Australian Statistical Geography Standard (ASGS). Under this classification, areas are defined as follows:

• Indigenous Locations (ILOCs): are aggregates of one or more SA1s. ILOCs generally represent small Aboriginal and Torres Strait Islander communities with a minimum population of 90 Aboriginal and Torres Strait Islander usual residents. An ILOC is an area designed to allow the production of Census statistics relating to Aboriginal and Torres Strait Islander people with a high level of spatial accuracy while maintaining the confidentiality of individuals. For the 2016 ASGS, 1115 ILOCs have been defined to cover the whole of geographic Australia.
• Indigenous Areas (IAREs): are medium sized geographical units designed to facilitate the release of more detailed statistics. IAREs provide a balance between spatial resolution and increased granularity of attribute data. They are created by aggregating one or more ILOCs. For the 2016 Census, 430 IAREs are defined to cover the whole of geographic Australia.
• Indigenous Regions (IREGs): are large geographical units loosely based on the former Aboriginal and Torres Strait Islander Commission boundaries. They are created by aggregating one or more IAREs. The greater population of IREGs enables the highest level of granularity of attribute data through greater cross classification of variables compared with IAREs and ILOCs. For the 2016 Census 58 IREGs are defined to cover the whole of geographic Australia. IREGs do not cross state or territory borders.
   

30 For further information see Australian Statistical Geography Standard (ASGS): Volume 2 - Indigenous Structure (cat. no. 1270.0.55.002).

31 LGAs are an ABS approximation of official gazetted Local Government boundaries as defined by each State and Territory Local Government Department. These approximated boundaries are constructed from the allocation of one or more whole Mesh Blocks. LGAs are good for understanding the characteristics of an individual LGA at a point in time.

32 The Census and Statistics Act, 1905 provides the authority for the ABS to collect statistical information, and requires that statistical output shall not be published or disseminated in a manner that is likely to enable the identification of a particular person or organisation. This requirement means that the ABS must take care that identifiable information about individual respondents cannot be derived from published data.

33 To protect confidentiality within this publication, some small cell values have been suppressed.

34 In addition, some Remoteness Areas in the states and territories have been combined to protect confidentiality when disaggregated by other characteristics, such as age and sex. In Victoria, Outer Regional Australia and Remote Australia have been combined to produce Rest of Victoria. In Tasmania, Remote Australia, and Very Remote Australia have been combined to produce Rest of Tasmania. Remoteness Areas are not available for the ACT in this product.

Related publications and references

35 Other ABS publications that may be of interest to users of this publication include:

• Australian Demographic Statistics (cat. no. 3101.0)
• Australian Historical Population Statistics (cat. no. 3105.0.65.001)
• Births, Australia (cat. no. 3301.0)
• Census Dictionary (cat. no. 2901.0)
• Census of Population and Housing - Counts of Aboriginal and Torres Strait Islander Australians (cat. no. 2075.0)
• Census of Population and Housing - Details of Overcount and Undercount, Australia (cat. no. 2940.0)
• Deaths, Australia (cat. no. 3302.0)
• Demography Working Paper 2001/4 - Issues in Estimating the Indigenous Population (cat. no. 3126.0)
• Life Tables for Aboriginal and Torres Strait Islander Australians (cat. no. 3302.0.55.003)
• Estimates and Projections, Aboriginal and Torres Strait Islander Australians (cat. no. 3238.0)
• Occasional Paper: Population Issues, Indigenous Australians (cat. no. 4708.0)
• Census of Population and Housing: Understanding the Increase in Aboriginal and Torres Strait Islander Counts Between 2011 and 2016 (cat. no. 2077.0) (to be released October 2018)
   

36 Current publications and other products released by the ABS are listed on the ABS web site www.abs.gov.au.

37 As well as the statistics included in this and related publications, additional information is available from the ABS web site at www.abs.gov.au and accessing Statistics/Population.

1 The final estimate of the Aboriginal and Torres Strait Islander population at 30 June 2016 was 798,400 people. This estimate comprises of 649,200 people counted on Census night, plus the 138,000 people measured as net undercount and the inclusion of an additional adjustment of 11,200 people. This technical note outlines the method by which the Australian Bureau of Statistics (ABS) used data from the 2016 Census of Population and Housing and Census Post Enumeration Survey (PES) to produce final rebased estimates of the Aboriginal and Torres Strait Islander population for 30 June 2016. Final Aboriginal and Torres Strait Islander estimates superseed the preliminary Aboriginal and Torres Strait Islander estimates released on 27 September 2017 in Australian Demographic Statistics, March 2017 (cat. no. 3101.0).

2 The Australian Bureau of Statistics (ABS) used the results of the 2016 Census of Population and Housing as the main data source to produce the rebased population estimates for Aboriginal and Torres Strait Islander Australians. The Census counted 649,200 Aboriginal and Torres Strait Islander people in Australia (including Other Territories) on Census night. This excludes 1,411,000 Census records (6.0% of the total Census count) with Indigenous status unknown. Of these, 14% or (0.9% of the total Census count) were a result of item non-response; that is, the ABS received a partially completed Census form for the person, with the Indigenous status question unanswered.

The remaining and majority (86%) of records with unknown Indigenous status (5.2% of the total Census count) for which no Census form was received and the dwelling (either private or non-private) was deemed to be occupied, then had people imputed into it. While some of the records will be for people of Aboriginal and/or Torres Strait Islander origin and others for non-Indigenous people, no imputation was made for Indigenous status in Census data. These Census counts were then adjusted using data collected in the Census Post Enumeration Survey.

3 The Census Post Enumeration Survey (PES) is a household survey conducted by the ABS shortly after each Census, that provides an independent measure of Census coverage. The 2016 PES included people from approximately 42,000 fully responding households across Australia. Information was collected for everyone present in the household. In addition to obtaining basic demographic information, questions were asked about each person's usual residence, their location on Census night, and any other addresses where they might have been counted in the Census. The PES results are used to determine how many people should have been counted in the Census, how many people were missed, how many were counted more than once, and how many were counted in error. Net undercount is defined as the difference between the PES estimate of the number of people who should have been counted in the Census and the actual Census count.

4 Net undercount is used as one of the inputs for compiling the Aboriginal and Torres Strait Islander population estimates. In the PES, Indigenous status is collected from every person (i.e. there is no non-response). The PES is therefore able to estimate the undercount of the Aboriginal and Torres Strait Islander population for those Census records for which Indigenous status was not stated (i.e. the undercount of the Aboriginal and Torres Strait Islander population in the Census due to non-response).

5 To estimate net undercount for the Aboriginal and Torres Strait Islander population, the PES imputed unknown Indigenous status where it was unknown in the Census. Specifically, a not stated Indigenous status was imputed as either Aboriginal and Torres Strait Islander or non-Indigenous, according to the distribution of stated responses within each age group, sex, Census form type and geographical area (Statistical Area Level 2) according to PES.

6 In the 2016 Census, the net undercount of the Aboriginal and Torres Strait Islander population was 138,000 people. This equates to a net undercount rate of 17.5%, which is slightly higher than the rate for the 2011 Census (17.2%). This undercount number is derived by calculating the difference between the PES Aboriginal and Torres Strait Islander population estimate (787,200) and the Aboriginal and Torres Strait Islander population Census count (649,200).

7 Some state and territory estimates of undercount for the Aboriginal and Torres Strait Islander population have relatively high standard errors and therefore the raw PES estimates are not used in calculating Aboriginal and Torres Strait Islander population estimates below the national level. Instead, the Empirical Bayes method (as used in 2011) is used to produce smoothed undercount rates for 18 regions of Australia. Each of these regions was a customised geographic area designed to capture the varied collection issues in different parts of Australia, as an input into the Empirical Bayesian estimation method. Victoria, Tasmania, South Australia and the Northern Territory were separated into two regions (split between capital city and balance of state), while New South Wales, Queensland and Western Australia were separated into three regions (with the balance of state split). The Australian Capital Territory was treated as a single region.

8 This method smooths the raw PES estimate of the undercount based on the Census characteristics of the region (specifically the proportion of Aboriginal and Torres Strait Islander persons and the level of Census non-response in the region).

9 The aim of smoothing is to provide stable estimates with lower standard errors. The overall amount of smoothing used in each region was determined by two factors; the standard error for each region and an overall smoothing constant. The smoothing constant was chosen using a ‘method of moments’ technique developed by Morris (1983). Regions with high standard errors required more smoothing. The outcome of this methodology, in relation to smaller standard errors and confidence intervals is presented in the table and the graph below.

10 Final estimates of the Aboriginal and Torres Strait Islander population on Census night, which have been adjusted for net undercount (as noted above) were then further adjusted to compile population estimates at 30 June. This involved:

• adding back Australian residents temporarily overseas (RTO) on Census night using actual travel behaviour (unlike preliminary RTO estimates that used the behaviour of travellers at the same time the previous year); and
• backdating from Census night to 30 June using components of population growth with finalised births, deaths and migration data (where as preliminary estimates use preliminary births and deaths data).
   

These additional adjustments added in 11,200 people at the national level.

11 The table below shows preliminary and final estimates as well as 2011 Census-based projections of the Aboriginal and Torres Strait Islander population, for 30 June 2016. For Australia, the final estimate remained largely unchanged from the preliminary estimate published on 27 September 2017 in Australian Demographic Statistics, March 2017 (cat. no. 3101.0).