The quality and availability of information on Aboriginal and Torres Strait Islander peoples has improved markedly over the last decade. This improvement is largely due to a more co-ordinated approach to information on Indigenous peoples in the health, community services and housing sectors, as well as regular reporting on the health and welfare of Indigenous peoples.
In addition, because Indigenous Australians are now counted more accurately through the Census, there is better administrative data. Specific surveys of Aboriginal and Torres Strait Islander people are also conducted regularly to fill gaps in health and welfare information.
Despite these achievements, many challenges remain. Varying levels of Indigenous identification between data collections, within data collections over time, and between regions, make it harder to assess changes over time and between regions. Mismatch between how people identify (or are identified) in Censuses compared to some administrative datasets-for example, death records-poses specific problems such as potential under-estimates of Indigenous morbidity. And collecting information that is culturally relevant to Aboriginal and Torres Strait Islander peoples can be challenging.
The quality and availability of Indigenous statistics are also affected by the statistical and practical challenges of surveying a population comprising just 2.5% of the total Australian population, especially when one-quarter of the Indigenous population lives in remote or very remote areas.
National standards
National standards ensure consistency and provide confidence that the same concept is being measured the same way across different data collections. National standards allow different data sets to be used together or compared.
National standards in health and welfare statistics are published in the National health data dictionary (NHDD) and on METeOR (AIHW's online metadata registry).
For Indigenous status, self-reporting is the most accurate way to gather this information. To ensure consistency and accuracy, there is a national standard for how the question is asked, and how the answers are recorded.
The national standard for Indigenous status is:
Indigenous data limitations
Inconsistencies in the Indigenous status question
The standard question, responses and recording categories for Aboriginal and Torres Strait Islander identification have been endorsed nationally but are not used in all data collections or across all jurisdictions. This limits the ability to compare the data produced.
Under-identification
For administrative data, the main limitation in Indigenous statistics is undercounting or under-identification. This happens when clients are not asked about their Indigenous status, the standard question is asked inconsistently, or recorded inaccurately. For example, a person's Indigenous status may be recorded as 'non-Indigenous' or 'not stated' (instead of 'No' 'Yes, Aboriginal' and/or 'Yes, Torres Strait Islander'). These incorrectly recorded responses are included in systems for monitoring and understanding the health and welfare of Aboriginal and Torres Strait Islander Australians so this 'incorrect' or 'missing' information raises problems for analysing the data and drawing conclusions. Because we don't know whether the characteristics of these 'missing' Aboriginal and Torres Strait Islander clients are different to those that have been reported as Aboriginal and/or Torres Strait Islander, we don't know whether the available data are biased.
The degree of undercounting or under-identification in data sets can vary across jurisdictions, sectors and service providers. This makes it difficult to measure the gap in health outcomes between Indigenous and non-Indigenous Australians, and to monitor progress in closing the gap.
Under-identification also creates difficulties in understanding Aboriginal and Torres Strait Islander people's preferences in health and welfare services, and which services and policies deliver the best outcomes. This limits researchers and policy-makers in understanding what works to overcome Indigenous disadvantage and improve health outcomes.
Small sample sizes
For survey data, smaller numbers of Indigenous Australians limits the quality and availability of Indigenous health and welfare data.
With the exception of selected national Indigenous specific surveys, most surveys which include a sample of Indigenous Australians have a greater likelihood of statistical errors as a result of the small sample size, which affects the ability to make comparisons across population groups, jurisdictions or over time.
Infrequent collection
Infrequent collection also limits the quality and availability of Indigenous health and welfare data.
Data sources such as the Census and National Aboriginal and Torres Strait Islander Health Survey occur only every five to six years, which limits the ability to measure changes in health outcomes and monitor annual trends.
National information on Indigenous Australians
The AIHW holds a number of national administrative data sets in health and community services. Within these data sets, the AIHW is investigating methods to estimate the level of Indigenous under-identification. This will provide valuable information on the quality of these collections, and may mean that future data can be corrected to adjust for under-identification.
Community Services data collections
- Aged Care data sets (various)
- Disability Services National Minimum Data Set
- Juvenile Justice National Minimum Data Set
- National Child Protection Data Collection
- Specialist Homelessness Services Data Collection (formally Supported Accommodation Assistance Program (SAAP) National Data Collection)
Read more in Aboriginal and Torres Strait Islander identification in community services data collections: an updated data quality report
Health data sets
Available through AIHW METeOR:
Census and population estimates
The Census of Population and Housing, conducted every five years, provides the basis for estimating the size of the Indigenous population. It also provides information on housing, employment, education, income and language, at all levels of geography. The range of Indigenous-specific information has been expanding and improving with each Census since 1971.
The most recent Census was conducted in August 2011 with results expected in 2013.
There are two main issues to consider when using Census data:
- The first is non-response to the Indigenous status question. In the 2006 Census, for example, there was a non-response rate of 1.7% for this question.
- The second issue is that the census form is not completed at the same time for everyone, and a small number of people may be counted more than once, or not at all.
The net effect of overcount and undercount is called net undercount. To measure this, the ABS conducts another survey (the post enumeration survey or PES) shortly after the Census. The PES provides an independent check of census coverage. Based on the PES, the ABS estimates that the net undercount for Aboriginal and Torres Strait Islander people was 11.5% in the 2006 Census, compared with 2.7% for the rest of the population (ABS 2007a).
Nevertheless, estimates of the size and characteristics of the Aboriginal and Torres Strait Islander population are derived from the Census. The estimated resident Aboriginal and Torres Strait Islander population at 30 June 2006 was 517,000. This represents an increase of 12% between the 2001 and 2006 censuses, while the census count for the total population increased by 6% over the same period. Preliminary estimates of the resident Aboriginal and Torres Strait Islander population at 30 June 2011 was 670,000 which represents an increase of 30% between the 2006 and 2011 censuses. Note that these numbers may differ to some degree from the final estimates which are due for release by the ABS in August 2013.
Following each Census, assumptions are made about past levels of mortality to produce back cast population estimates and forward projected population estimates for the Indigenous population. These experimental estimates and projections are used extensively in estimating a range of health indicators, and in some cases estimating the extent of under-identification in health data collections.
Indigenous-specific survey data
Three national Indigenous-specific surveys are conducted on five to six yearly cycles to help understand the social environment and health and wellbeing of the Indigenous population. The surveys ensure that data are relevant, that collection methods are sensitive to Aboriginal and Torres Strait Islander peoples, and that broad consultation occurs.
1. The National Aboriginal and Torres Strait Islander Social Survey (NATSISS)
The NATSISS, conducted by the ABS every six years, provides broad information across areas of social importance, including health, housing, work, education and income. The results allow relationships between different areas of social concern to be explored and help us understand the extent to which some people face multiple social disadvantages.
The latest NATSISS was conducted in 2008. Information was collected by personal interview from about 13,300 Aboriginal and Torres Strait Islander Australians, 5,500 of whom were aged 0-14 years and 7,800 aged 15 years and over. Interviews were conducted throughout Australia, including remote areas.
2. The National Aboriginal and Torres Strait Islander Health Survey (NATSIHS)
The NATSIHS was first conducted by the ABS in 2004-05, and is conducted approximately every six years to coincide with the National Health Survey. The information collected in the survey includes health status, risk factors and actions and socio-economic circumstances.
The next NATSIHS is being conducted during 2012-13 as part of the Australian Health Survey which will for the first time include detailed nutrition, physical activity and biomedical information to provide the most comprehensive snapshot to-date of Australia's health.
The 2004-05 NATSIHS included 10,439 Aboriginal and Torres Strait Islander Australians, which was considerably larger than the 2001 National Health Survey (which included a sample of 3,681 Aboriginal and Torres Strait Islander Australians).
The NATSIHS was specifically designed to represent Aboriginal and Torres Strait Islander Australians. Information recorded in this survey is 'as reported' by respondents. The survey was conducted in private dwellings in both remote and non-remote regions across Australia.
3. Community Housing and Infrastructure Needs Survey (CHINS)
The CHINS collects data about discrete Indigenous communities including housing and management information from Aboriginal and Torres Strait Islander housing organisations, and a range of community infrastructure information for locations identified as discrete communities.
CHINS data include details of the current housing stock, and management practices and financial arrangements of Indigenous housing organisations. Information is also collected on housing (including water, power and sewerage), and other education and health services.
The CHINS survey was first conducted in 1999 and subsequent surveys were conducted in 2001 and 2006 in conjunction with the Census.
In 2006, a total of 1,187 communities were surveyed, with a combined usual population of 92,960 Aboriginal and Torres Strait Islander peoples or 18% of the total estimated Indigenous population.
Read more: Housing and Infrastructure in Aboriginal and Torres Strait Islander Communities, Australia, 2006
Data development
The Australian Government is committed to improving the availability of good quality Indigenous health data. It actively engages with key stakeholders, state and territory governments, the ABS and the AIHW to improve Indigenous health statistics through the National Advisory Group on Aboriginal and Torres Strait Islander Health Information Development (NAGATSIHID).
Some COAG initiatives will further improve Indigenous data quality. The National Indigenous Reform Agreement (COAG 2008) strongly emphasises the need to close the gap by improving data quality across the health, education and employment sectors.
The AIHW, ABS and state and territory governments are improving the quality of data by:
- improving Indigenous counting in the Census
- capturing data related to smoking, and alcohol use during pregnancy, antenatal care and Indigenous status of the baby (previously, only Indigenous status of the mother has been recorded)
- analysing the level of Indigenous identification in key datasets
- implementing and evaluating best practice guidelines for the collection of Indigenous status in health data collections
- promoting national best practice guidelines for data linkage relevant to Indigenous identification
- developing a national Key Performance Indicator (KPI) framework for Indigenous primary health care services.
Read more: National Indigenous Reform Agreement (Schedule F).
National agreements
National Advisory Group on Aboriginal and Torres Strait Islander Health Information Strategic Plan 2011-2015
Further information