COMPONENTS OF NET UNDERCOUNT
While net undercount estimates measure the completeness of Census counts, and adjustment factors can be used to adjust Census counts, it is also useful to understand the contributing components of net undercount.
For the purposes of PES estimation, persons are categorised into the contact sector or the non-contact sector.
The contact sector includes:
- persons in dwellings for which a Census form was received before the commencement of PES enumeration (which includes persons overcounted and persons missed from these forms)
- persons from occupied dwellings that were entirely missed by the Census
- persons missed by Census because their dwellings were mistakenly deemed unoccupied on Census night.
The non-contact sector
- persons in dwellings deemed occupied on Census night, from which no Census form was received. Data for these persons were imputed during Census processing
- persons in dwellings whose Census form was received after the commencement of PES enumeration (i.e. a late return)
- persons with insufficient personal identifier information on their Census form.
The PES traditionally measures a net undercount of persons in the contact sector, with the number of persons missing from Census forms (including persons whose dwelling was missed entirely) generally exceeding the number of persons counted multiple times or in error. In contrast, the non-contact sector is typically characterised by an overcount of persons, which is essentially a measure of over-imputation for non-responding dwellings in the Census that were deemed occupied.
These trends hold true for the 2016 PES.
Components of Net Undercount, 2011-2016
(a) Persons missing from completed Census forms and persons in dwellings missed by Census (i.e. dwellings not known to Census) or dwellings mistakenly deemed unoccupied.
(b) Persons included on Census forms multiple times or included in error.
(c) A negative value indicates a net overcount.
(d) Net overcount for the non-contact sector represents over-imputation. This column includes a small contribution from late returns and Census records with insufficient personal identifier information.
(e) A positive difference as the net overcount in the non-contact sector was higher in 2016 compared with 2011.
(f) Percentages are calculated using the total PES population estimate as the denominator.
1 150 588
The large positive difference in gross overcount in the contact sector and in net overcount in the non-contact sector in 2016, compared with 2011 (changes of 90.3% and 77.2%, respectively), are driving the lower total net undercount at the Australia level. Specifically, the larger changes in overcount (including over-imputation in the Census) have offset, to a large degree, the gross undercount in the contact sector. The positive difference for the latter changed by much less in 2016 (a change of 30.0%), compared with the overcount, thus driving the total net undercount down.
Note that net undercount for the contact sector at the Australia level is equal to gross undercount minus gross overcount. This is not true for estimates below the national level, where differences in response categories between the Census and PES, and not stated values in the Census, are also taken into account.
For a detailed discussion on the sub-components that comprise both the contact and the non-contact sectors, see Contact Sector
and Non-contact Sector
on the Summary tab.