• Summary
• Downloads
• Explanatory Notes
• Related Information
• Past Releases

SAMPLE DESIGN

In the PES, private dwellings (houses, flats, etc.) and discrete Indigenous communities are separately identified and sampled. In total, about 35,000 dwellings are expected to respond to the PES, including approximately 500 dwellings selected from 28 discrete Indigenous communities.

Table 1 shows the expected number of fully responding households in the private dwelling sample and discrete Indigenous community sample for each State and Territory. It is important to note that, due to sample loss and other non-response, the number of dwellings selected in the sample will be greater than shown below. Based on the results of other ABS surveys, the rate of sample loss is expected to vary between different components of the sample, which is taken into account.

Table 1: Expected number of fully responding households
	Private dwellings	ICF dwellings(a)	Total
	no.	no.	no.
New South Wales	7 618	0	7 618
Victoria	6 450	0	6 450
Queensland	6 394	48 (3)	6 442
South Australia	3 694	18 (2)	3 712
Western Australia	4 410	33 (5)	4 443
Tasmania	1 971	0	1 971
Northern Territory	2 460	382 (18)	2 842
Australian Capital Territory	1 349	0	1 349
Australia	34 346	481 (28)	34 827

• males and females at Australia level - less than 0.2 percentage points;
• males and females by age group at Australia level - less than 0.6 percentage points;
• persons by state/territory - less than 0.6 percentage points (except for the Northern Territory, where it is expected to be less than 1.5 percentage points); and
• Indigenous people at Australia level - less than 2.5 percentage points.

The mainstream sample of private dwellings is derived from the 2006 Monthly Population Survey Parallel Sample frame, which is based on the Statistical Division and Subdivision structure of the Australian Standard Geographical Classification (ASGC). This sampling framework first divides Australia into 100 geographical areas. These areas are then divided into strata according to population density, remoteness and growth, then:

• in the first stage of selection, a sample of Census collection districts (CDs) is selected (systematically, with probability proportional to size) to represent each stratum;
• in the second stage of selection, each selected CD is divided into smaller areas called blocks, one of which is selected (again systematically, with probability proportional to size); and
• in the third stage, a sample of dwellings in the selected block is taken using systematic equal probability sampling.

Discrete Indigenous communities sample

CDs with an identified Indigenous population of greater than 75% are classified differently from the remaining population for both operational and sampling reasons. These CDs and the Indigenous communities within them form the Indigenous Communities frame (ICF).

For the purposes of sampling, discrete Indigenous communities are grouped into 'sets' comprising main communities and their associated outstations. The selection of main communities is undertaken with probability proportional to the size of the set. The aim is to select as representative a sample as possible while also considering cost constraints, reasonable interviewer workloads and expected sample size.

If a community is selected in the sample, selection of dwellings within the community follows the same procedure as for the selection of private dwellings within selected blocks in the non-ICF component of the sample. That is, an interviewer compiles a list of all the dwellings within the community and a sample is taken using systematic equal probability sampling.

A selection of outstations associated with each selected main community is also included in the sample. Each outstation has an equal chance of selection and, once selected, all dwellings within the outstation are enumerated.

Sampling strategies to improve Indigenous population estimates

Initial attempts to improve the quality of Indigenous undercount information occurred in 2006, when remote areas and discrete Indigenous communities were enumerated for the first time. The 2011 PES strives to further improve the estimates of Indigenous undercount by increasing the number of discrete Indigenous communities selected, as well as the total number of dwellings selected within the community sample. This increase in sample is expected to decrease the sampling error on estimates of Indigenous undercount.

In addition to the increase in the discrete Indigenous communities sample, the 2011 PES will increase the number of selected dwellings from areas outside the discrete Indigenous communities that have been identified as having a relatively high proportion of Indigenous persons. This focused sampling of additional areas within the mainstream sample is a key sample design change from 2006.

The focused sampling approach will involve a combination of extra dwellings selected from within the parallel sample, as well as some additional selections from CDs outside the parallel sample.

This focus on obtaining an increase in Indigenous respondents within the mainstream sample is of particular importance, given approximately three-quarters of the Indigenous population usually reside outside of communities.

The impact of the transition to the ASGS

With the geographic classification used in the Census changing from the Australian Standard Geographical Classification (ASGC) to the Australian Statistical Geography Standard (ASGS), there will be a mixture of geographies used in the 2011 PES, with different geographic units used at different stages of the survey cycle. This development presents challenges for the ABS to manage in relation to the sample design.

As in 2006, the 2011 PES sample design will be structured by CDs. This is in contrast to the 2011 PES field operations, processing and output, which will use geographical units from the ASGS, primarily Statistical Area Level 1 (SA1). To overcome this challenge, the PES sample will be recoded to the ASGS geographies prior to the start of PES enumeration.

The second challenge posed is the spatial definition of discrete Indigenous communities, some of which may be different between the two geography frameworks. This will be managed by a mapping exercise to compare the two geographies for the 28 communities selected in the PES sample.