2940.0.55.002 - Information Paper: Measuring Overcount and Undercount in the 2016 Population Census, Jul 2016  
ARCHIVED ISSUE Released at 11:30 AM (CANBERRA TIME) 01/07/2016  First Issue
   Page tools: Print Print Page Print all pages in this productPrint All


Initially, a count of Australian residents who were in Australia on Census night is obtained from the Census. This Census count is adjusted on the basis of net overcount or net undercount, to account for people being counted more than once and people being missed in the Census. The basis of the estimate of net overcount or net undercount is the PES.

ERP calculations then make an adjustment for Australian residents who were temporarily overseas on Census night. An estimate of this number is made using data from completed passenger cards, visa and passport information obtained from the Department of Immigration and Citizenship and these people are added into the Australian resident population.

The final step in calculating ERP is to backdate to 30 June of the Census year. This is achieved by adding the deaths and subtracting the births and net overseas migration which occurred between 1 July and the Census date. Table 2 shows the components used to calculate ERP for Australia from the 2011 Census.

Table 2. Components of ERP, Australia, 2011

Components of ERP (a)

Census count, actual location
21 727.2
less Overseas visitors
Census count, place of usual residence
21 507.7
plus Net undercount (b)
plus Residents temporarily overseas
ERP (a) as at 9 August 2011
22 378.8
less Births (1 July to 9 August 2011 )
plus Deaths (1 July to 9 August 2011 )
less Net overseas migration (1 July to 9 August 2011)
ERP (a) as at 30 June 2011
22 340.0

(a) Preliminary ERP.
(b) Includes demographic adjustments.

Information on the calculation of the ERP for 30 June 2016 based on the 2016 Census will be reported in Australian Demographic Statistics, December quarter 2016 (cat. no. 3101.0), due for release on 21 June 2017. Additional information can also be found in Population Estimates: Concepts, Sources and Methods, 2009 (cat. no. 3228.0.55.001).


While the PES identifies people and dwellings missed in the Census, the extent to which some people are missed in both the Census and the PES may not be fully reflected in PES estimation, which would result in a correlation bias. As in any survey, the PES is also subject to sampling and non-sampling error.

To offset the impact of correlation bias and survey error, population estimates derived from the PES are further refined using demographic adjustments based on three sources of independent population information: the National Demographic Data Bank, Medicare enrolment numbers, and the estimated resident population based on the previous Census. These sources have different strengths and weaknesses, but where the data are considered to be most reliable, they are used for comparison with PES adjusted age and sex population distributions, and for possible minor adjustments to population estimates.

The National Demographic Data Bank is a population database maintained by the ABS using administrative data (notably births, deaths, and overseas arrivals and departures). The database is independent of Census data and contains population data back to 1925. For the 2006 PES, these data were considered to measure age-sex totals well up to about age 35, after which there were some concerns about pre-1970 international migration data. Sex ratios derived from these data are considered most reliable for ages under 28 years.

Enrolment data from Medicare (the Australian government health rebate system) are considered a good source for calculating sex ratios, but less reliable for age-sex population totals. Age-sex totals are least reliable among the older ages where people may remain enrolled in Medicare after their death until Medicare Australia is notified and the record updated.