Background and Documentation
ABOUT MORTALITY DATA FOR AUSTRALIA
Modified from the documentation in HMD by Vladimir Canudas-Romo and Wen Su
Early population statistics for Australia stem from custom population enumerations known as musters. The musters served specific administrative purposes and targeted particular groups of populations, such as convicts. They were held irregularly over the first 40 years of settlement with a total count of musters that was no more than 30. People were not legally required to participate. As a result, such enumerations suffer from a great undercount (Camm, 1984).
Population statistics for Australia in the period covered by the AHMD derive mainly from national censuses. The first modern census was conducted in 1828, but this only covered New South Wales and parts of Tasmania and Queensland. Since that time censuses were held more or less regularly every 5 or 10 years in the different colonies. Starting in 1911, following the establishment of the Federation of Colonies on 1 January 1901, censuses were conducted regularly by the successive governments of the Commonwealth of Australia. A list of Australian censuses is provided in Appendix Table I.
The first records of vital statistics are available from parish registers, established after colonization began in 1788. Official registration was first established in Tasmania in the year 1838 followed by Western Australia, South Australia, Victoria and New South Wales including Queensland in the years 1841, 1842, 1853, and 1856, respectively (Young, 1975). Before 1906, data on deaths are available from the annual State vital statistical reports; from 1907 to 1963, they are from the statistical reports of the Australian Bureau of Statistics (ABS) and its predecessor (Commonwealth Bureau of Census and Statistics 1905–1974). However, population estimates have been available for the entire nation for every calendar year starting in 1921 only. The ABS now maintains an individual record database of deaths starting in 1964.
Source of Data
Population data for years 1971-2020 are official mid-year estimates from the ABS. Whenever possible, the latest population estimates are included. These data have been downloaded freely from the ABS website.
Data on deaths for years 1971 to 2020 have been compiled by the AHMD from annual official statistics reports.
Annual birth counts by sex have also been downloaded from the ABS website. Additional information about the data sources is included in Appendix II. Complete information for each data value in the database is provided in the reference and note files, available in the input database.
There has been no change in territorial coverage of Australian statistics in the period currently covered by the AHMD series. The estimated resident population of Australia includes the population of “Other Territories” since June 1994, but it is not treated as a territorial change in this database.
Until 1967, the First Australians population was excluded from federal censuses and population estimates as stipulated in section 127 of the Constitution. ‘Aboriginal natives’ were defined as individuals with more than 50% First Australians’ ‘blood’. The Constitution Alteration (Aboriginals) Act of 1967 repealed this provision and since 10 August 1967 ‘full-blooded’ Indigenous and Torres Strait Islander people are officially included in population statistics. Official population estimates were backdated to 1961 to include estimates of Indigenous population, but First Australians’ births and deaths have only been included in Australia's vital statistics since 1966. The ABS estimates the First Australians’ population to have been approximately 80,000 persons at the time of the 1966 census, around 0.7-0.8% of the total population.
DEATH COUNT DATA
Coverage and Completeness
Due to the legal requirements of the death registration system established in the middle of the 19th century, data on deaths are considered to be complete and of a good quality though data collected after 1964 are considered to be of a better quality than historical data (HMD communication with ABS). Deaths of overseas visitors are classified according to the State or Territory where the death was registered. Official death statistics disseminated by ABS are based on deaths tabulated by year of registration rather than year of occurrence. In recent years, about five percent of deaths are registered after the year of occurrence. The effect of late registration would have introduced a random error in life expectancy on the order of +/- 0.1 year of life, but with no detectable bias. Most late-registered deaths are registered in the year following occurrence. For example, the most recent year in the AHMD series, 2019, incorporates deaths which occurred in 2019 but were registered in 2020, and is considered reasonably complete. Forthcoming updates will tend to lag behind preliminary data availability in this fashion.
Currently, death statistics come from the death certificates and the death information forms. Death certificates are filled by the attending medical officer and include information on the cause of death or the condition leading to death. A death information form is filled out by the deceased’s next of kin or by the funeral director and includes administrative and demographic information. Death certificates are coded by ABS and added to the unit record level database.
POPULATION COUNT DATA
Coverage and Completeness
Prior to 1971, population estimates are based on de facto census counts. Starting in 1971, population estimates are based on de jure census counts adjusted for net undercount and to include Australians temporarily overseas. Data are considered to be complete and of good quality.
Resident (de jure) population estimates are referred to as either Preliminary, Revised or Final. They are derived from the census date population estimates which are in turn de jure census counts adjusted for net undercount. The net census undercount is estimated by ABS from a Post-Enumeration Survey conducted 3 weeks after the census. The estimates are produced by the cohort component method using information on subsequent births, deaths and migration. Preliminary estimates are generally made available six months after the reference period. Revised estimates are generally published a year later (e.g. estimates for the 1997-98 financial year became available in September, 1999). Final population estimates are published for the previous intercensal period after each census. Estimates by single years of age and sex are published annually with a June 30th reference date.
- All estimates from 1971 to 2020 are official mid-year estimates based on final intercensal estimates by ABS.
- The estimates for the years 2016-2021 are the revised official estimates, based on intercensal estimation from the 2021 census.
BIRTH COUNT DATA
Coverage and Completeness
Due to the legal requirements of the birth registration system established in the middle of the 19th century, data on births are considered to be complete and of good quality. Annual births are included in the AHMD series by year of registration and sex. The AHMD birth series begins in 1971.
There is usually an interval between the occurrence and registration of a birth and, as a result, some births occurring in one year are not registered until the following year or later, especially those taking place at the very end of a calendar year. This can be caused by either a delay by the parent(s) in submitting a completed form to the registry, or a delay by the registry in processing the birth. (Australian Bureau of Statistics, 2022).
Australian Bureau of Statistics. (2004). ”Australian Historical Population Statistics”.
Catalogue number 3105.0.065.001. Australian Bureau of Statistics.
Australian Bureau of Statistics. (2004). “3228.0 Demographic Estimates and Projections: Concepts, Sources and Methods”. (Statistical Concepts Library, 2004).
Australian Bureau of Statistics (2022). “3301.0 - Births, Australia, 2015 – Explanatory Notes”. Accessed Dec. 5, 2016 from http://www.abs.gov.au/ausstats/[email protected]/exnote/3301.0.
Australian Institute of Health and Welfare. (2004). National Mortality Database Documentation. Australian Institute of Health and Welfare.
Australian National University. (2005). Australian Demographic Databank, Version
3.2b. Australian Centre for Population Research.
Camm, J. (1984). “Past Population of Australia: A Review of the Historical Development of Australian Colonial Censuses, 1828-1901”. Research Papers in Geography No. 28. New South Wales: University of Newcastle.
Commonwealth Bureau of Census and Statistics (1947), Canberra, Australia.
Demography 1945, Bulletin No. 63
Forster, C. and C. Hazlehurst. (1988). Australian Statisticians and the Development of official Statistics. In Year Book Australia, Catalogue number 1301.0. Canberra: Australian Bureau of Statistic.
HMD. Human Mortality Database. Max Planck Institute for Demographic Research (Germany), University of California, Berkeley (USA), and French Institute for Demographic Studies (France). Available at www.mortality.org.
Kippen, Rebecca and Lucas, David. (2004). “Sources for Australian Historical Demography “. Working Papers in Demography, No. 93. Demography and Sociology Program. Research School of Social Sciences. The Australian National University.
Trewin, D. (2000). 2001 Census of Population and Housing. “How Australia Takes a Census”. Catalogue no. 2903.0 Australian Bureau of Statistics. Inline refernce E:\Data\mDb\Australia\Trewin.ABS.2000.pdf
Young, C. (1975). “Epidemic of infectious diseases in Australia prior to 1914”. Colloquium on Crises in Australia of Mortality, University of Montreal.