This study contains time series of data of the Annual Aboriginal Census for Australia, Australian Capital Territory, New South Wales, Northern Territory, Queensland, South Australia, Tasmania, Victoria and Western Australia from 1921 to 1944. The file au.edu.anu.ada.ddi.20002-NSW is the data for New South Wales.

Special care notice:
Aboriginal and Torres Strait Islander people, researchers and other users should be aware that material in this dataset may contain material that is considered offensive. The data has been retained in its original format because it represents an evidential record of language, beliefs or other cultural situations at a point in time.


The Annual Aboriginal Census is considered as a significant official source of Aboriginal population statistics. It was conducted annually in June from 1921 to 1944, exempting the war years between 1941 and 1944 in each State and Territory. The 1944 census was incomplete with New South Wales not taking part at all. Enumeration of Aboriginal populations was poor and difficulties in classification occurred. The Census was a collaboration of the Commonwealth Bureau of Census and Statistics who initiated the study, State and Territory Statisticians, the Protector of Aborigines, and local police officers who conducted the enumeration. The Annual Aboriginal Census is also referred to as the Annual Census of Aborigines and Police Census.

The enumeration for the Annual Aboriginal Census was conducted by Local Police officers within their Police District. This was conducted with no self-identification by those enumerated but classified by the local police officers making the census, a population summary with no unit record data collected. Due to the frequent use of rounded number figures have been suggested to be rough estimates.

The Conference of Statisticians, Sydney, 1925 resolved that the future Annual Aboriginal Census would include; number and distribution of 'full-bloods', number and distribution of 'half-castes' living with 'full-bloods', males and females, adults and children, and whether nomadic, in regular employment or in protected camps. Changes to classification also occurred at the Statisticians' Conference in Perth in 1926. It was resolved that 'persons of mixed blood living with aboriginals should be classed as 'half-caste' aboriginals, whatever be the degree of the white strain' and that 'persons of mixed blood not living with aboriginals shall be 'half-caste' if strains are approximately equal, as 'full blood' if the predominant strain is aboriginal, and not included at all if the predominant strain is white'. Previous to this no definition was giving for determination of 'half-caste' enumeration. The Statistician's Conference in Hobart in 1928 furthered the classification of social and employment status, children of persons in employ were classed as 'other' and those in both 'regular employ' and 'in supervised camps' would be classed only as in 'regular employ'. In 1939 the Queensland statistician insisted the counting of Torres Strait Islander people be separated from Aboriginal counts.

Leonard Smith, obtained Annual Aboriginal Census state level data with his work with the National Population Inquiry and as a part of his PH.D thesis. An ARC grant in conjunction with Gordon Briscoe allowed local area data to be obtained. Data came in format of copies of the original collection forms, tabulation sheets and published bulletins of results. The data was obtained from the Australian Bureau of Statistics, and the Commonwealth, State and Territory Archives.