3302.0.55.003 - Life Tables for Aboriginal and Torres Strait Islander Australians, 2010-2012  
ARCHIVED ISSUE Released at 11:30 AM (CANBERRA TIME) 15/11/2013   
   Page tools: Print Print Page Print all pages in this productPrint All  
Contents >> Quality issues with Aboriginal and Torres Strait Islander deaths and population data >> Population estimates of Aboriginal and Torres Strait Islander Australians

POPULATION ESTIMATES OF ABORIGINAL AND TORRES STRAIT ISLANDER AUSTRALIANS

The second component necessary for the compilation of life tables for Aboriginal and Torres Strait Islander Australians is information on the relevant population exposed to the risk of dying; that is, the population of Aboriginal and Torres Strait Islander Australians, by age and sex. These are required as denominators in the calculation of age-specific death rates from which life tables are produced.


Changes in the Aboriginal and Torres Strait Islander population, 2006-2011

The Census of Population and Housing is the principal source of information about Australia's population. It has been held on a five-yearly basis since 1961, with the most recent conducted in August 2011.

The Census provides the foundation from which Australia's official population figures-estimated resident population (ERP)-is calculated. The Census count of the population is adjusted for:

  • estimates of the number of people missed in the Census;
  • estimates of those counted more than once in the Census;
  • temporary visitors from overseas;
  • Australian residents temporarily overseas (RTO) on Census night; and
  • backdating from Census night to the ERP reference date of 30 June of the Census year using data on births, deaths, and interstate and overseas migration.

This process results in an estimate of Australia's total population (by age and sex) as at 30 June of the Census year. For intercensal years (that is, years other than the Census year), this Census-based ERP is incremented by adding births and net overseas migration and subtracting deaths. However, estimating the size and composition of the Aboriginal and Torres Strait Islander population is more complicated. In intercensal years this standard approach cannot be used due to the lack of sufficiently reliable data on Aboriginal and Torres Strait Islander births, deaths and migration.

For the five-yearly Census, there are a number of issues that make compilation of Aboriginal and Torres Strait Islander ERP problematic. These include:
  • undercount of the Aboriginal and Torres Strait Islander population;
  • non-response to the Indigenous status question on the Census form; and
  • unexplained growth in Aboriginal and Torres Strait Islander Census counts relative to the previous Census.

Undercount of Aboriginal and Torres Strait Islander population

While every effort is made to ensure full coverage of people and dwellings in the Census, some people are missed (undercount) and others are counted more than once (overcount). In Australia, more people are missed from the Census than are counted more than once. The net effect of overcount and undercount is called net undercount. The ABS conducts a Post Enumeration Survey (PES) about one month after each Census to measure the extent of net undercount in the Census. Estimates of net undercount provide direct information on the quality of population counts in the Census, and enable the necessary adjustment or correction to be made to the raw Census counts.

In addition, for some people, the Indigenous status reported in the PES is different to the Indigenous status recorded in the Census. Accordingly, estimates from the PES include an adjustment for misclassification error.

The net undercount rate in the 2011 Census was estimated at 17.2% for the Aboriginal and Torres Strait Islander population, compared to an undercount of 6.2% for the non-Indigenous population. Corresponding estimates for the 2006 Census were 11.5% for the Aboriginal and Torres Strait Islander population and 8.2% for the non-Indigenous population.

However, it is important to note that the net undercount rates are not strictly comparable over time due to changes in both Census and PES methodologies. The 2011 PES improved the collection of Indigenous status and also utilised the new methodology of Automated Data Linking, which resulted in better linking and matching of PES and Census records, and a better measure of net undercount. For more details on the PES and Aboriginal and Torres Strait Islander undercount, see:
Non-response to Indigenous status question on the Census form

Despite the best efforts of the Australian Bureau of Statistics (ABS) to collect complete information from all people in Australia on Census night, there will always be a group of people for whom Indigenous status is not known. While some people with an unknown Indigenous status will be Aboriginal and/or Torres Strait Islander people, and some will be non-Indigenous people, the exact proportions cannot be determined from the Census data. This complicates analysis of Census counts, given Indigenous status is not imputed within Census information, but is instead imputed later within the PES for use in producing population estimates (imputation is a statistical process for predicting values where no response was provided to a question and a response could not be derived).

The number of Census records with an unknown Indigenous status in the 2011 Census was 1,058,600 (5% of the total Census count) compared with 1,133,400 records (6% of the total Census count) in 2006. The difference of 74,800 records represents a 7% decrease in records with an unknown Indigenous status between the 2006 and 2011 Censuses. This means that more people had a known Indigenous status in 2011 than in 2006, and this may account for some of the 'unexplained' increase in Census counts of Aboriginal and Torres Strait Islander people between 2006 and 2011.

The Northern Territory had the highest proportion of records with an unknown Indigenous status in 2011 (8%), followed by Western Australia (6%) and Queensland and New South Wales (both 5%).

There are two broad situations which result in a person's Indigenous status being unknown:
  • the returned Census form does not have a response to the Indigenous status question (known as item non-response), or
  • when no Census form is received from a dwelling the number of males and females and other key demographics variables are imputed. Indigenous status is not imputed for these records, but is coded to 'not stated'.

While some of these records will be for people of Aboriginal and Torres Strait Islander origin and the others for non-Indigenous origin, no imputation was made for Indigenous status in the Census file. However, for compiling Aboriginal and Torres Strait Islander ERP they are allocated as either Aboriginal and Torres Strait Islander or non-Indigenous according to the distribution of stated responses within each geographic area, age group, sex and Census form type. For more details on not stated Indigenous status and how Aboriginal and Torres Strait Islander ERP are derived, see Estimates of Aboriginal and Torres Strait Islander Australians, Jun 2011 (cat. no. 3238.0.55.001) and Census of Population and Housing: Understanding the Increase in Aboriginal and Torres Strait Islander Counts, 2006-2011 (cat. no. 2077.0).

Unexplained increase in Aboriginal and Torres Strait Islander Census counts

The way the Indigenous status of a person is recorded in the Census can change over time and in different situations. The recorded status of an individual may move between any of the categories of Indigenous status, including between Aboriginal and Torres Strait Islander and non-Indigenous. This is known as a change in a person's propensity to identify.

Over the past 35 years, there has been a clear upward trend in Aboriginal and Torres Strait Islander Census counts, beginning with the 1971 Census and continuing to the 2011 Census. During this time, large increases in Aboriginal and Torres Strait Islander Census counts have occurred on several occasions. Natural increase (the excess of births over deaths) can account for a proportion, but not all, of these increases, while overseas migration has had an insignificant effect on the size of the Aboriginal and Torres Strait Islander population. It therefore appears that a further component of change exists in regard to Aboriginal and Torres Strait Islander Census counts, referred to as 'error of closure'.

In recent times, there has been particularly large increases between two intercensal periods, 1991-1996 and 2006-2011. Between 1991 and 1996, the Census count of Aboriginal and Torres Strait Islander people increased by 88,000 (33%). The components of this increase were estimated to be 14% due to natural increase, and the remaining 19% due to other factors, including changes in Census procedures and a difference in the identification of people in the Census as being of Aboriginal and Torres Strait Islander origin.

Between 2006 and 2011, the Census count of Aboriginal and Torres Strait Islander people increased by 93,300 (21%). The components of this increase were estimated to be 14% due to natural increase, and the remaining 7% due to other factors, including changes in Census procedures as well as a difference in the identification of people in the Census as being of Aboriginal and Torres Strait Islander origin. For more information on the change in Census counts of Aboriginal and Torres Strait Islander people from 2006 to 2011 see Census of Population and Housing: Understanding the Increase in Aboriginal and Torres Strait Islander Counts, 2006-2011 (cat. no. 2077.0).

Age/sex structure

For the purposes of compiling life tables it is necessary to have accurate measures of the population according to their age and sex. It is therefore important to make some assessment of the quality of the age/sex structure of Aboriginal and Torres Strait Islander ERP for Australia and the states and territories. The age distributions of the 30 June 2006 and 30 June 2011 Aboriginal and Torres Strait Islander ERP have remained more or less the same (graphs 2.4 and 2.5).

2.4 Age structure, Australia - Aboriginal and Torres Strait Islander males
Graph: 4.4 Age structure, Australia—Aboriginal and Torres Strait Islander males


2.5 Age structure, Australia - Aboriginal and Torres Strait Islander females
Graph: 4.5 Age structure, Australia—Aboriginal and Torres Strait Islander females


In terms of the variation in the age distribution between successive Census year ERPs, the index of dissimilarity (Shryock et al., 1976:131) showed quite low values in comparison to its theoretical range of 0 to 100. The smaller the index of dissimilarity values, the more similar are the two age distributions. For the states and territories, most index of dissimilarity values fall around or below 5 except for the Australian Capital Territory, where the index value could be unreliable due to the small size of the Aboriginal and Torres Strait Islander population (table 2.6).

2.6 INDEX OF DISSIMILARITY BETWEEN ABORIGINAL AND TORRES STRAIT ISLANDER AGE STRUCTURES - 30 June 2006 and 30 June 2011

Males
Females
State/territory
index
index

NSW
3.5
3.9
Vic.
3.7
3.6
Qld
3.2
2.8
SA
3.9
4.0
WA
3.2
2.9
Tas.
4.5
5.0
NT
4.0
3.2
ACT
9.7
8.1
Aust.(a)
3.3
3.1

(a) Includes Other Territories.








Previous PageNext Page