3228.0.55.001 - Population Estimates: Concepts, Sources and Methods, 2009  
ARCHIVED ISSUE Released at 11:30 AM (CANBERRA TIME) 12/06/2009   
   Page tools: Print Print Page Print all pages in this productPrint All

APPENDIX 1 DEMOGRAPHIC ADJUSTMENT


INTRODUCTION

A1.1 Interim population estimates as calculated using the method specified in Chapter 2 - Estimating national and state population, are subject to adjustments based on demographic analysis ('demographic adjustments'). Most of these demographic adjustments (DAs) are made to the interim estimates at the Australian level, on the basis of a number of checks against other sources (data confrontation) as well as internal checks on both Census and undercount from the Census Post Enumeration Survey (PES).

A1.2 Although most DAs are one-off adjustments, some are regularly required, these include:

  • cohort size and sex ratio check versus other sources
  • age/sex internal consistency for undercount
  • consistency against other Census years' undercount adjustments
  • checks of ages imputed in the Census
  • checks for Census 'faux-old' (formerly 'vandals') in upper ages
  • assorted other issues that arise during processing, including SLA-specific adjustments.

A1.3 The overall approach to making DAs is to make conservative, minimally intrusive adjustments.


DATA CONFRONTATION

A1.4 The data confrontation process compares the population and sex ratio in each age group (both single year and 5-yearly age groups) of this interim estimate with the same data from a range of other sources. These include:
  • Demographic estimates derived from the National Demographic Databank (see below). This represents the total population best in older ages and sex ratios in younger ages (ages less than about 28 years).
  • Medicare registrations. The scope is close to the Australian resident population, so this represents population size and sex ratios well.
  • ERP for the current Census year based on the previous Census. Note however that the ERP based on the previous Census is sometimes correlated with the other two sources due to adjustments made on the previous Census. Used for comparing both population counts and sex ratios.
  • Current Census counts by usual residence. These are useful for showing what raw Census data gives before PES, RTOs, backdating, etc. are applied. Used for comparing both population counts and sex ratios.
  • ERP for the previous Census year, offset 5 years. As it excludes components of population change over the last intercensal period, this is mainly used for sex ratio analysis - particularly if there are concerns about the reliability of intercensal net overseas migration (NOM).
  • Previous Census counts by usual residence, offset 5 years. Mainly used for sex ratio analysis.
  • AEC enrolments. Not as useful for comparison with population counts as the scope is narrower than the Australian resident population, but useful for sex ratio analysis.
  • Births. The population count and sex ratio of 0-year olds is compared to those of births over the year preceding the Census.

A1.5 Prior to 2006, there appeared to be a clear pattern in which the numbers of people aged 30, 40, 50, 60 and 70 were slightly high, and the numbers of people aged 31, 41, 51, 61 and 71 were slightly low in comparison with the independent sources. As a result of this 'age heaping', the population in these pairs of consecutive ages were combined and then redistributed according to the Demographic Databank age distribution.

A1.6 In the 2006 Census, respondents were given the option of providing their age in years or giving date of birth. The majority chose to answer the date of birth option, which is not easy to generalise, therefore in 2006 no adjustment was required to correct age heaping.


UNDERCOUNT CONSISTENCY

A1.7 The PES undercount rate is checked for internal consistency by comparing age/sex undercount rates with the four previous PES-based rates, comparing the male with the female undercount rate, and comparing the undercount rate by sex in successive age groups.

A1.8 It is worth noting that the PES is designed to adjust for missing and erroneous Census reporting, but can never hope to perfectly do so. The reasons for this limitation include:
  • correlation bias of PES and Census non-response (i.e. the reasons which contribute to a person being missed in the Census may also cause them to be missed in the PES), see correlation bias in the Glossary for more information
  • issues of matching/searching to non-contact dwellings (there is no corresponding person in the Census for the PES to match to)
  • sample size (PES estimates of the population in the non-contact sector have relatively high sampling error because of the small sample size)
  • non-universal scope of PES (people included in the Census that were not able to be sampled in the PES, e.g. homeless people, people in non-private dwellings such as hotels, motels, hospitals, etc.).

A1.9 For more information see:

MAGNITUDE OF DEMOGRAPHIC ADJUSTMENTS

A1.10 For final rebasing following the 2006 Census, the demographic adjustments netted to -14,300 or 0.07% of the population (-15,200 in preliminary rebasing), and in effect alter the population as defined by the PES estimates. This compares with net adjustments of -24,900 and -8,700 for preliminary and final rebasing respectively after the 2001 Census. To set some perspective, note that PES net undercount adjustment totals 550,000 and RTOs are 346,000. See the table below for the effect of DAs.

A1.1 Adjustment components of estimated resident population, Final - 30 June 2006

NSW
Vic.
Qld
SA
WA
Tas.
NT
ACT
Aust.(a)
'000
'000
'000
'000
'000
'000
'000
'000
'000

Persons
Components as at 8 August 2006:
Census count, actual location
6 585.7
4 915.3
4 046.9
1 509.0
1 986.2
470.8
217.1
327.9
20 061.6
plus - Residents absent interstate
74.8
87.7
38.1
28.4
19.2
13.0
4.5
10.5
276.3
less - Interstate visitors
54.3
36.2
107.5
13.8
25.1
5.5
22.1
11.4
276.3
less - Overseas visitors
57.0
34.4
72.9
9.2
21.3
1.8
6.7
2.9
206.4
equals - Census count, place of usual residence
6 549.2
4 932.4
3 904.5
1 514.3
1 959.1
476.5
192.9
324.0
19 855.3
plus - Allowance for under-enumeration
157.6
113.6
148.4
36.3
64.2
9.5
15.9
4.0
549.6
plus - Demographic adjustment
-5.1
-3.4
-2.8
-1.0
-1.3
-0.3
-0.2
-0.2
-14.3
plus - Residents temporarily overseas
124.3
94.6
50.5
20.2
42.5
4.6
2.5
6.8
346.0
equals - ERP as at 8 August 2006
6 826.0
5 137.3
4 100.6
1 569.8
2 064.4
490.3
211.1
334.7
20 736.6
Backdating components to 30 June 2006:
less - Births
9.6
7.1
6.0
2.0
2.9
0.7
0.4
0.5
29.3
plus - Deaths
5.4
3.9
2.9
1.4
1.3
0.4
0.1
0.2
15.6
less - Net interstate migration
-2.4
-0.1
2.4
-0.2
0.4
-0.1
0.1
-
-
less - Net overseas migration
8.2
7.6
4.2
1.5
3.1
0.1
0.1
0.2
25.0
equals - final ERP 30 June 2006
6 816.1
5 126.5
4 090.9
1 567.9
2 059.4
490.0
210.6
334.1
20 697.9

- nil or rounded to zero (including null cells)
(a) Includes Other Territories



NATIONAL DEMOGRAPHIC DATABANK

A1.11 In the 1960s, before the PES was fully operational, a computer file was created by H.P.Brown of the Australian National University (ANU) to enable an assessment of Census undercounting and age mis-statement by demographic analysis. Called the ANU Demographic Databank, this file contained a time-series of national demographic events (births, deaths and overseas arrivals and departures) and Census counts of population for all years back to 1921. The data, which were by age and sex, were obtained from the ABS, as were additional data used by A. R. Hall of the ANU to update the file for the period 30 June 1966 to 30 June 1976 (Hall 1980).

A1.12 The purpose of this file was to construct population counts that were independent of the Census by by-passing censuses and the adjustments made from time to time to conform to Census counts of the population. For example, it was possible to estimate the population at 30 June 2006 by commencing from the 1921 Census count and, using the cohort component method, updating only for subsequent demographic events (i.e. births, deaths and migrations). Any Census could be used as the base, and, once chosen, could be updated for any required number of years. In particular, though, these independent populations were compiled for Census dates to enable comparison with the Census count at that date.

A1.13 For the years since 1976, the Databank has been maintained and updated by the ABS, and is now called the National Demographic Databank. It is a useful supplement to the PES as a measure of the quality of Census data.

A1.14 Revisions have been made to the database to conform to the changes, as from 1971, in the definition of 'population', i.e. defining population as the number of people who usually reside in Australia (see Chapter 1 - Overview). These revisions entailed the exclusion of all short-term overseas arrivals and departures back to 1925. (It was not possible to go back to 1921 because data for short-term movements were not available for years prior to 1925).

A1.15 Although it would have been possible to maintain the old series as well as the revised one, it was decided that only the revised series would be maintained and updated for future use.

A1.16 For analysing 2006 Census undercount, an age range of 0 to 80 years could be derived for the population estimate from the Databank without using the 1921 Census as the base (since all such persons would have been born after 1921). This age range was considered sufficient for the purposes of comparison with the interim population estimates, while the PES alone was available for the higher age groups. The problem of not having short-term migration data before 1925 is thus not applicable for these ages.