1996 Census Undercount
The 1996 Census was held on Tuesday 6 August 1996. While the objective of the Census was to include everyone in Australia on this night (except diplomats and their families), inevitably some people were missed.
A survey was conducted after the census to find out how many people were missed. The results of this survey are available in the publication 1996 Census Data Quality: Undercount (Cat. No. 2940.0) released recently. The survey showed that the proportion of people missed was very small - an estimated 1.6% of people present in Australia on census night 1996 were not counted in the Census. This means that 98.4% of people who should have been counted in the Census were counted.
The data released from the 1996 Census (and other Censuses) are not adjusted for undercount. Users of census data should be aware that the data are subject to undercount and, in particular, that the level of undercount varies for different demographic groups.
Estimates of undercount are used in the estimation of the resident population of Australia. To produce the population estimates, census counts are also adjusted for undercount and other factors. Estimates of the resident population of Australia, the States and Territories are available in Australian Demographic Statistics (Cat. No. 3101.0).
The publication (1996 Census Data Quality: Undercount) provides information on the proportion and characteristics of people and dwellings missed in the census. Some key findings include:
- the undercount rate decreased from 1.8% in 1991 to 1.6% in 1996,
- undercount was highest in the Northern Territory (3.3%) and lowest in South Australia (1.2%),
- young adult males had a high undercount rate (4.3% for males aged 20 to 24) while seniors had relatively low rates (1% or less or people aged 55 or over),
- people born in New Zealand had a higher undercount rate (at 2.8%) than people born in Australia or other countries.
This page last updated 30 July 2007