3228.0 - Demographic Estimates and Projections: Concepts, Sources and Methods, 1999
ARCHIVED ISSUE Released at 11:30 AM (CANBERRA TIME) 30/08/1999
Page tools: Print All
Contents >> Chapter 5. Household estimates

Introduction

5.1. Annual household estimates for 30 June of each year are published by state, by capital city / balance of state and by household size and age (number of persons aged 0 to 14 and number of persons aged 15 or over). For the purposes of household estimates, households can be generally considered analogous to occupied private dwellings (OPDs).

5.2. The base populations from which subsequent annual estimates are derived are provided by the census, and are called census date household estimates. The method used to update these household estimates is a modified version of the household size propensity method outlined by D.S. Ironmonger and C.W. Lloyd-Smith and described in 'Projections of Households and Household Populations by Household Size Propensities', as published in the Journal of the Australian Population Association (1992, Vol 9, No 2).

5.3. Household estimates were first published in Household Estimates, Australia 1986, 1991-94 (3229.0), which also describes the method used to produce household estimates prior to 30 June 1996. This chapter focuses on the method used to produce household estimates since 30 June 1996. Annual estimates for 30 June are provided in Australian Demographic Statistics (3101.0).

5.4. Estimating the number and distribution of households and the household population involved three major steps. These were:

• preparation of a distribution of the household population by age, sex, State and part of State;
• calculation of household size propensities by age, sex, State and part of state (using Census data for Census date estimates and the Labour Force Survey for post-censal estimates); and
• application of the household size propensities to the household population to generate the number and distribution of households by size.

The following two sections discuss how census date estimates and the subsequent post-censal estimates are derived.

Census date household estimates (CDHE)

5.5. Household estimates for 30 June in the Census year are calculated at the SLA level by household size, using the following steps:

Step 1 - CDHE calculations.

5.6. Census counts of usual residents enumerated at home are compiled for each SLA by age (0 to 14 years, 15 years and over) and whether each person lived in an occupied private dwelling (OPD) or non-private dwelling (NPD). These counts are then used to estimate at the SLA level the percentage of people living in OPDs for each of the two age groups.

5.7. The restriction to those persons enumerated at home is used to exclude visitors and absentees to obtain a more accurate estimate of the percentage of persons living in OPDs.

Step 2 - CDHE calculations.

5.8. Multiply the percentage of people living in OPDs for the two age groups by their respective SLA estimated resident populations. This provides an estimate of the number of persons living in OPDs.

Step 3 - CDHE calculations.

5.9. Compile census counts of OPDs (ie. dwellings, not persons) by SLA of enumeration by the number of persons in each OPD aged 15 years or over and the number of persons aged less than 15 years (see Table 5.1). The number of persons (by age) is calculated by adding the number of usual residents at home on census night (by age) to the number of persons temporarily absent on census night (by age). As such visitors are excluded and people temporarily absent are included.

 5.1: Number of Occupied Private Dwellings by Household Composition Number of persons aged 15 years and over Number of persons aged less than 15 0 1 2 3 4 5 6+ 1 2 3 4 5 6+

Step 4 - CDHE calculations.

5.10. For each SLA, calculate the household population by household composition by age. This is done by multiplying the number of households for a given household composition by the number of persons in that age group (calculated in Step 3) for that household composition. For example, the number of persons aged 15 and over in households with 2 persons aged 15 and over and no persons aged less than 15 is equal to twice the number of households with 2 persons aged 15 and over and no persons aged less than 15.

5.11. For household compositions of "6 or more persons" (by age), the number of persons used to multiply by the household count is equal to the average household size of households with 6 or more persons (for the corresponding age group).

Step 5 - CDHE calculations.

5.12. Separately prorate the census date household population by household composition calculated in Step 4 by age such that the total household population matches the 30 June estimate of the number of persons living in OPDs previously calculated in Step 2.

Step 6 - CDHE calculations.

5.13. Estimate the number of households by household composition through dividing the number of total persons (summed over age from Step 5) by the household size. For example, the number of households with 1 person aged 15 years and over and 3 persons aged less than 15 would be equal to the total number of persons living in these households divided by 4.

5.14. Note that visitor only households and non-classifiable households are excluded from census date dwelling counts (approximately 214,000 households on 1996), although the population living in non-classifiable households (along with persons missed in the census and residents temporarily overseas) is re-included in Step 5 by prorating to the 30 June estimated resident population.

Post-censal household estimates (PCHE)

5.15. Using the census date (ie. 30 June) household estimates as the base population, post-censal household estimates are compiled in accordance with estimated resident population data and data obtained on household size propensities (by household composition) from the monthly Labour Force Survey (LFS).

The Labour Force Survey

5.16. The LFS is based on a multi-stage area sample of private dwellings and a list of non-private dwellings and covers about 0.5% of all persons aged 15 years and over who are usual residents of Australia. The survey excludes from scope members of the permanent Australian defence force, non-Australian diplomats, non-Australian diplomatic staff and non-Australian members of their families and members of non-Australian defence forces and their dependents.

5.17. The sample is divided into eight panels or groups of households called rotation groups. Each rotation group is surveyed each month for eight months and is then dropped from the sample. The outgoing rotation group is replaced by a new rotation group which then remains in the sample for eight months.

5.18. For the purposes of estimating household size propensities, data are pooled over a five-month period. All eight rotation groups are used for the relevant month along with two outgoing rotation groups from each of the two previous months and two incoming rotation groups from each of the two following months. This method of pooling increased the sample size by 50%, thus strengthening the stability of the estimates. The method of pooling also ensures that all households had only one chance of being included in the pooled data.

5.19. Post-censal household estimates are calculated using the following steps:

Step 1 - PCHE calculations.

5.20. Estimate the proportion of persons living in private dwellings, separately according to State, part of State, age and sex by extrapolating the proportion of persons living in private dwellings as calculated using data from the two previous censuses. For example, for post-censal estimates for 1997 to 2001:

 where D = Proportion of persons living in private dwellings q = Number of quarters since the June 1996 quarter = Proportion of persons living in private dwellings at the 1996 Census = Proportion of persons living in private dwellings at the 1991 Census

Step 2 - PCHE calculations.

5.21. Estimate the number of persons living in private dwellings, separately according to State, part of State, age and sex by multiplying the estimated resident population by its corresponding proportion of people living in private dwellings (as derived in Step 1 - PCHE calculations).

Step 3 - PCHE calculations.

5.22. Using pooled data from the LFS (as described in paragraph 5.18.), calculate (for each age, sex, State and part of State combination) the probability that such a person will live in a given household composition. These probabilities are referred to as household size propensities (Table 5.2).

 5.2: Household Size Propensities (a) by Household Composition Number of persons aged 15 years and over Number of persons aged less than 15 0 1 2 3+ 1 PLFS(1, 0) PLFS(1, 1) PLFS(1, 2) PLFS(1, 3+) 2 PLFS(2, 0) PLFS(2, 1) PLFS(2, 2) PLFS(2, 3+) 3 PLFS(3, 0) PLFS(3, 1) PLFS(3, 2) PLFS(3, 3+) 4+ PLFS(4+, 0) PLFS(4+, 1) PLFS(4+, 2) PLFS(4+, 3+) (a) The household propensity PLFS(i,j) is the probability of a person living in a private dwelling of "i" persons aged 15 and over and "j" persons aged less than 15.

Step 4 - PCHE calculations.
5.23. The Census provides the best basis for accurate estimates of the number of households in Australia. As household size propensities derived from LFS data for Census dates do not match Census-based household size propensities, the LFS-based household size propensities are adjusted to align with the Census-based household size propensities:
 P(i,j) = where P(i,j) = The adjusted household size propensity PLFS(i,j) = The household size propensity derived from the LFS PC,96(i,j) = The household size propensity derived from the Census for census date PLFS,96(i,j) = The household size propensity derived from the LFS for census date

5.24. The adjusted household size propensities are re-scaled so they sum to one.

Step 5 - PCHE calculations.

5.25. For each State by part of State, age and sex combination, calculate the number of persons living in OPDs by household composition by proportionally distributing the estimated resident population living in OPDs (from Step 2 - PCHE calculations) according to the corresponding adjusted household size propensities (calculated in Step 4 - PCHE calculations).

Step 6 - PCHE calculations.

5.26. For each State by part of State, calculate the number of persons aged 15 or over living in OPDs by household composition by summing the person estimates from Step 5 - PCHE calculations over sex and the corresponding age groups.

5.27. The number of households of a given household composition equals the number of persons aged 15 or over divided by the number aged 15 or over in the given household composition. For example, the number of households with 2 persons aged 15 or over is equal to half the number of people aged 15 and over who live in households with 2 persons aged 15 or over.

5.28. Total households are then derived by summing across all household compositions thus:

 H = where H = number of households N15+ (i) = number of persons aged 15 and over living in private dwellings containing persons aged 15 and over.

5.29. As was the case in 5.11, for households of 4 or more persons aged 15 and over, the number of persons used is equal to the average number of persons aged 15 and over living in such households.

 Previous Page Next Page