1 This publication contains experimental estimates of the Aboriginal and Torres Strait Islander (Indigenous) population of Australia, from 1991 to 2001, based on the 2001 census. It also contains experimental projections of the Indigenous population for the period 2002 to 2009, based on the 2001 census. Descriptions of the methods used to produce the estimates and projections assumptions are included.
2 The Indigenous population is comprised of people who are of Aboriginal origin, Torres Strait Islander origin or both Aboriginal and Torres Strait Islander origin. The Torres Strait Islander estimates comprise people of Torres Strait Islander origin only and people of both Aboriginal and Torres Strait Islander origin. The Torres Strait Islander only estimates comprise people who are of Torres Strait Islander origin only.
3 The estimates and projections of the Aboriginal and Torres Strait Islander (Indigenous) population presented in this publication are experimental. The significant intercensal volatility in Indigenous census counts and the quality of the data on births, deaths and migration do not support the standard approach to population estimation.
4 Estimates and projections for Jervis Bay Territory, Christmas Island and Cocos (Keeling) Islands are included in the totals for Australia. These territories are collectively referred to as 'Other Territories' in this publication.
Source of statistics
5 The 2001 census data, on which the estimates and projections are based, are compiled on a usual residence basis. In the census, place of usual residence refers to the address at which a person has lived or intends to live for a total of six months or more in the census year. Persons who do not report a place of usual residence, or are of indeterminate address or homeless, are imputed as being a resident of their place of enumeration.
Three stage process
6 The method of estimation was a three stage process. Firstly, various adjustments were applied to census counts of Indigenous persons by Statistical Local Area (SLA) of usual residence to produce the estimated population as at 7 August 2001. Secondly, the effects of births, deaths, overseas migration and interstate migration for the 38 days between 1 July and 7 August 2001 were removed to derive 30 June 2001 estimates. Finally, these 30 June 2001 estimates were 'survived' back to 30 June 1991 using life tables. This method is known as the reverse survival method. Zero net internal and external migration were assumed for backdating the 30 June 2001 estimates to 30 June 1991.
7 Various adjustments were made to the 7 August 2001 estimates. The adjustment process takes account of these factors: non-response to the Aboriginal and Torres Strait Islander origin question in the census; unknown Indigenous status on census forms imputed by ABS when a form could not be obtained from persons identified in the field; net census undercount of Indigenous persons; and residents temporarily overseas (RTOs) on census night.
Not stated Indigenous origin
8 The first adjustment dealt with completed census forms for which there was no response to the Indigenous origin question for some of the people reported on the form. The second adjustment was made in instances where the census collector considered that a person did reside at a particular residence on census night, but was not able to obtain a completed census form for that person. In this case, a 'substitute' census form was created where age, sex and the SLA of usual residence on census night were imputed, but all other characteristics (including Aboriginal and/or Torres Strait Islander origin) were recorded as 'Not Stated'.
9 It is assumed that some of the people for whom the Indigenous origin question was not answered, or for which substitute forms were created, were of Aboriginal and/or Torres Strait Islander origin. Therefore the not stated census counts for Indigenous status were redistributed pro rata according to the reported Indigenous status proportions by SLA, age, sex and census form-type.
10 The third adjustment involved correcting for net undercount of the Indigenous population. Indigenous and non-Indigenous undercount rates were calculated for Australia using information obtained from the 2001 Post Enumeration Survey (PES). The following table shows the net Indigenous and non-Indigenous undercount rates, standard errors (SEs) and relative standard errors (RSEs) from the 2001 PES.
11 The Australian net undercount rate was used with census counts to obtain a census population adjusted for net undercount, called PES estimates. State by age and sex PES estimates were then derived from Australia by age and sex, and state by age PES estimates.
Net undercount estimate, SE and RSE by Indigenous status, 2001 PES
12 State by age, sex and Indigenous PES estimates were calculated using the state by age, sex and Indigenous census counts and the Indigenous net undercount rate and then constraining to state by age and sex PES estimates. The Indigenous PES estimates were then divided by the unconstrained data to obtain state by age, sex and Indigenous undercount adjustment factors. The undercount adjustment factors were applied to SLA Indigenous census counts by age and sex to derive Indigenous synthetic PES estimates at the SLA by age and sex level. These age structures were then backdated to 30 June 2001.
13 Two sets of adjustments were calculated for each SLA. Firstly, a correction for census over-imputation of persons in non-contact dwellings, and secondly an adjustment to allow for fully imputed persons (system created records) in private dwellings to be non-usual residents of the SLA. These adjustments were combined with the residents temporarily overseas (RTOs) on census night, births, deaths, internal migration and overseas migration backdating components.
14 The RTO backdating component was based on address information obtained from a large sample of all passenger cards. The addresses of RTOs were keyed in and then coded to SLAs to produce a state age/sex RTO distribution which was then applied.
15 Births and deaths occurring between 1 July 2001 and 7 August 2001 (census night) were obtained from ABS' regular processing of registered events, with usual address coded to SLA. The births data were adjusted slightly for historically observed registration lags.
16 Backdating estimates for overseas and internal migration were derived from the 2001 census data on usual residence one year ago. These were constrained to state-level backdating components for overseas arrivals and departures and net interstate migration respectively. One year ago overseas departure data are not available from the census, therefore SLA arrivals data were used instead to distribute state level overseas departures.
17 Through estimating Indigenous and non-Indigenous population concurrently while constraining to Capital City and Balance of State ERP and SLA synthetic-ERP, RTOs and backdating components which are not available by Indigenous status are implicitly apportioned pro rata.
18 Finally the SLA by age and sex experimental Indigenous estimated resident population (ERP) was derived using the SLA by age and sex Indigenous PES synthetic estimates which were adjusted for census imputations, RTOs and backdating components. Indigenous ERP for higher geographic areas were then derived by aggregating the SLA level ERPs.
19 After deriving experimental Indigenous resident population estimates as at 30 June 2001, the estimates were survived back one year at a time to 30 June 1991 using a reverse survival technique based on experimental Indigenous life tables. This was done because the quality of the component series (census counts, births and deaths) do not support the application of the cohort component method. This series is useful for studying growth in the Indigenous population in the period 1991-2001. However, caution should be exercised in calculating birth rates and death rates using these backdated population figures as Indigenous population estimates are likely to be inconsistent with Indigenous identification in the births and deaths collections.
20 The absolute size of net internal migration does not warrant a specific interstate migration assumption in constructing experimental estimates. As such, zero internal migration has been assumed. Census data also indicates that the level of external migration of Indigenous persons is negligible. Hence zero external migration has also been assumed.
21 Experimental Indigenous life tables for Australia are shown in the Appendix. An explanation of the method used to derive the life tables is given in paragraph 30 of the Explanatory Notes. The Indigenous mortality level observed during the 1996-2001 period was used to backcast the population estimates from 2001 to 1991.
22 There are many methods which may be used for population projections, ranging from simple extrapolations through broad economic, social and time series analyses to detailed component methods. The choice of method arguably depends on the resources at hand and the availability of reliable data.
23 The ABS uses the cohort component method, which begins with a base population for each sex by single year of age and advances it year by year, by applying assumptions regarding future fertility, mortality, migration and unexplained growth. This procedure is repeated for each year in the projection period for each state and territory and for Australia. The resulting population projections for each year for the states and territories, by sex and single years of age, are adjusted to sum to the Australian results.
24 The method used to generate the fertility assumption was a two stage process. Firstly, registered births data by age of mother (in the case of births to Indigenous women) and age of father (in the case of births to non-Indigenous women and Indigenous men), and the estimated Indigenous population were combined to produce preliminary age-specific fertility (in the case of births to Indigenous women) and paternity rates (in the case of births to non-Indigenous women and Indigenous men).
25 Secondly, these preliminary fertility and paternity rates were multiplied by a factor to adjust for undercoverage of registered births. Both fertility and paternity rates were increased by the same factor separately at the state and territory level.
Preliminary age-specific rates
26 Registered Indigenous births for the calendar years 2000 to 2002 were split into births to Indigenous mothers and births to Indigenous fathers and non-Indigenous mothers, by five-year age group, by state and territory. The preliminary age-specific fertility rates for each state and territory were calculated by dividing the average annual number of births to Indigenous mothers registered during 2000-2002 by the June 2001 experimental Indigenous female population estimate. Similarly, the preliminary age-specific paternity rates for each state and territory were calculated by dividing the average annual number of births to Indigenous fathers and non-Indigenous mothers registered during 2000-2002 by the June 2001 experimental Indigenous male population estimate. Births were averaged over this three-year period to smooth out the irregularities from year to year in the number of births, bearing in mind the relatively small number of Indigenous births in some areas and age groups.
Preliminary age-specific fertility rates of Indigenous mothers(a) - 30 June 2001
AGE GROUP (YEARS)
|State or territory|
Total fertility rate(b)
|New South Wales|
|Australian Capital Territory|
|- nil or rounded to zero (including null cells)|
|(a) Per thousand female population.|
|(b) The sum of five-year age-specific fertility rates (live births at each age of mother per female population of that age) multiplied by five.|
|(c) Includes Other Territories.|
Preliminary age-specific paternity rates of Indigenous fathers(a) - 30 June 2001
AGE GROUP (YEARS)
|State or territory|
Total paternity rate(b)
|New South Wales|
|Australian Capital Territory|
|(a) Per thousand male population.|
|(b) The sum of five-year age-specific paternity rates (live births at each age of father per male population at that age) multiplied by five.|
|(c) Includes Other Territories.|
Calculation of factors
27 The number of births that would be generated if the estimated 30 June 2001 Indigenous population experienced the preliminary fertility and paternity rates were calculated for each state and territory. The expected number of births was taken as the average of the estimated Indigenous population at age zero, one and two as at 30 June 2001. A factor was then calculated by dividing the expected births by the preliminary births.
Calculated births using preliminary fertility and paternity rates - 30 June 2001
|State or territory|
Births generated using preliminary rates
|New South Wales|
|Australian Capital Territory|
|(a) Includes Other Territories.|
28 For each state and territory, the preliminary age-specific fertility and paternity rates were then multiplied by the factor calculated above. The resultant age-specific rates were then used as the basis for the fertility assumption in the projections.
29 The standard approach to calculating death rates is to divide the number of deaths in a given population by the 'exposed to risk' population during that period. Although it is considered likely that most Indigenous births and deaths are registered, a proportion of these births and deaths are not identified as ‘Indigenous’. Without complete and accurate data on Indigenous births and deaths, and accurate data on the size and structure of the Indigenous population, the standard method for calculating Indigenous death rates cannot be used.
30 ABS used a new demographic method for determining consistency factors to make registered intercensal Indigenous deaths data consistent with two end census date experimental Indigenous population estimates. This method was developed by Bhat (2002) and offers a definite improvement over other indirect methods available for estimating mortality from incomplete data. The main advantage of this method over other methods is that this technique explicitly allows for an adjustment for migration to be taken into account. Using this method, and providing for unexplained population growth as the migration variable, the ABS calculates the consistency of Indigenous death registrations for the 1996-2001 intercensal period relative to the population estimates at the beginning and end of the five year period. The consistency factor, assumed to be constant at each age group, is applied to the observed age-specific death rates which are then used for the calculation of the Indigenous life tables. For further information, see the ABS Demography Working Paper 2004/3 - Calculating Life Tables for Use in Population Estimates and Projections of Aboriginal and Torres Strait Islander Australians (ABS cat. no. 3106.0.55.003). Experimental Indigenous life tables for Australia are shown in the appendix.
31 Recent censuses have shown that the in-migration rate of Indigenous persons is very low. It is assumed that out-migration is similarly negligible. As such, nil overseas migration with zero arrivals and zero departures was assumed for the projections.
32 The method used to generate the interstate migration assumption was a two stage process. First, the net census interstate flows for the 1996-2001 period were calculated for each state and territory and sex. Then these net flows were adjusted for differences between Indigenous population estimates and Indigenous census counts.
33 The projections used interstate arrival and departure rates for Indigenous people which were calculated separately for each state and territory, single year of age and sex using the 2001 census distribution of internal movements. These interstate arrival and departure rates were then constrained to net interstate flows.
Adjusting for differences between population estimates and census counts
34 The reliability of Indigenous data from the census is affected by a number of factors including census undercount and not stated Indigenous origin. To produce a better assumption for net interstate migration, both arrival and departure flows were multiplied by a proportion calculated by dividing Indigenous population estimates by Indigenous census counts. This calculation was made for each sex and state and territory.
35 The intercensal growth in the Indigenous population counts cannot be fully explained by births, deaths and migration. One way to measure this unexplained growth is for an earlier census date population estimate to be survived to the next census date and the difference between the next census date population estimate and the survived estimate remains unexplained.
36 Unexplained growth in the Indigenous population between the 1996 and 2001 censuses was estimated in a three stage process. First, the 30 June 1996 experimental Indigenous population estimate was survived to 30 June 2001 using the 1996-2001 experimental Indigenous life tables produced using the Bhat method. Secondly, the survived estimate was subtracted from the 30 June 2001 experimental Indigenous population estimate to get total unexplained growth in the Indigenous population during the 1996-2001 intercensal period. Finally, this total unexplained growth in the population was divided by the 30 June 1996 experimental Indigenous estimate and then expressed as an average annual growth rate. This calculation was made for each state and territory and sex.
Aboriginal and Torres Strait Islander origin
37 The 2001 census asked the following question of each person:
38 The question appeared as question 17 in the 2001 census forms (and as question 14 in the 1996 census). It was identical in both the censuses.
39 The census question is based on the first aspect of the definition of Aboriginal and Torres Strait Islander origin, adopted by the Commonwealth Government in 1978, but may, in part, measure the second aspect as well. The Commonwealth definition states that an Aboriginal or Torres Strait Islander person is:
40 Although the question on the census form itself included the instruction: "For persons of both Aboriginal and Torres Strait Islander origin, mark both 'Yes' boxes", some assistance was available in the 2001 Census Guide, which was distributed with each census form, and through the Census Inquiry Service. The Census Guide provided the following information:
- a person of Aboriginal or Torres Strait Islander descent
- who identifies as an Aboriginal or Torres Strait Islander
- is accepted as such by the community in which he or she lives.
"'Torres Strait Islander' origin refers to people who come from the Torres Strait Islands located between the Australian mainland and Papua New Guinea."
41 The Census Inquiry Service provided some additional information:
"For census purposes, 'Aboriginal' does not refer to everyone born in Australia, or to the aboriginal people of any other country.
Mark 'Yes, Aboriginal' if you are of Australian Aboriginal descent and identify yourself as Aboriginal.
'Torres Strait Islander' origin refers to people who come from the Torres Strait Islands located between the Australian mainland and Papua New Guinea.
Mark 'Yes, Torres Strait Islander' if you are of Torres Strait Islander descent and identify yourself as a Torres Strait Islander.
If you are of both Aboriginal and Torres Strait Islander origins you should mark both 'Yes' boxes.
If you are a descendant of South Sea Islanders or from another group of islands, you should mark 'No', unless you have an ancestor who was an Aboriginal person or from the islands between the Australian mainland and Papua New Guinea".
42 Number of births to Indigenous women per 1,000 Indigenous women.
43 The age at which half the population is older and half is younger.
44 Excess of births over deaths.
45 Number of births to non-Indigenous women and Indigenous men per 1,000 Indigenous men.
46 Male population per 100 female population.
Total fertility rate
47 The sum of age-specific fertility rates (live births at each age of mother per female population of that age). It represents the number of children a woman would bear during her lifetime if she experienced current age-specific fertility rates at each age of her reproductive life.
RELIABILITY OF PROJECTION RESULTS
48 The projection results published by the ABS are not intended as predictions or forecasts, but are illustrations of growth and change in the population which would occur if the assumptions about future demographic trends prevail over the projection period.
49 While the assumptions for the projections are formulated on the basis of an assessment of past trends, there is no certainty that any of the assumptions will or will not be realised over the projection period.
50 Accordingly, alternative projection series have been generated in recognition of this uncertainty and to provide users with a range of options.
51 Sex ratios which are based on a small population base are subject to high variability and should be interpreted with caution.
RELATED PUBLICATIONS AND REFERENCES
52 Other ABS publications that may be of interest to users of this publication include:
53 Current publications and other products released by the ABS are listed in the Catalogue of Publications and Products (cat. no. 1101.0). This catalogue is available from any ABS office or the ABS web site <http://www.abs.gov.au>. The ABS also issues a daily Release Advice on the web site which details products to be released in the week ahead.
- Australian Demographics Statistics (cat. no. 3101.0) - issued quarterly
- Births, Australia (cat. no. 3301.0) - issued annually
- Census of Population and Housing: Population Growth and Distribution, Australia, 2001 (cat. no. 2035.0)
- Census of Population and Housing: Aboriginal and Torres Strait Islander People, Australia, 1996 (cat. no. 2034.0)
- Deaths, Australia (cat. no. 3302.0) - issued annually
- Demography Working Paper 2001/4 - Issues in Estimating the Indigenous Population
- Demography Working Paper 2004/3 - Calculating Experimental Life Tables for Use in Population Estimates and Projections of Aboriginal and Torres Strait Islander Australians (cat. no. 3106.0.55.003)
- Experimental Estimates of the Aboriginal and Torres Strait islander Population, 1991 to 1996 (cat. no. 3230.0)
- Experimental Projections of the Aboriginal and Torres Strait Islander Population, 1996 to 2006 (cat. no. 3231.0)
- National Aboriginal and Torres Strait Islander Social Survey, 2002 (cat. no. 4714.0)
- Population Characteristics, Aboriginal and Torres Strait Islander Australians, 2001 (cat. no. 4713.0)
- Population Distribution, Indigenous Australians, 2001 (cat. no. 4705.0)
- Ross, Kate 1999. Occasional Paper: Population Issues, Indigenous Australians (cat. no. 4708.0)
- Aboriginal and Torres Strait Islander Population. Year Book, Australia, 2004 edition (cat. no. 1301.0)
- The Health and Welfare of Australia's Aboriginal and Torres Strait Islander Peoples, 2003 (cat. no. 4704.0)
54 As well as the statistics included in this and related publications, additional information is available from the ABS web site at <http://www.abs.gov.au> and accessing Themes/Demography.
55 Related publications and articles which may also be of interest are:
- Bhat, P.N. Mari (2002) General growth balance method: a reformulation for populations open to migration, Population Studies, 56, pp. 23-34. London, England.
- Preston, S.H. and K.J. Hill. 1980. 'Estimating the completeness of death registration', Population Studies, Vol. 34, no. 2, pp. 349-366.
ADDITIONAL STATISTICS AVAILABLE
56 A consultancy service to undertake Indigenous population estimates and projections under client-specified assumptions is available. For further information or a quote, contact Shahidullah, Demography Section, Canberra, on (02) 6252 5129.