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

2016 PES SAMPLE AND RESPONSE RATES

PES SAMPLE DESIGN

1 The 2016 PES sample comprised two components: private dwellings (houses, apartments, etc.) within the General Population sample, and private dwellings within the Discrete Community sample.

2 A total of 53,800 dwellings were randomly selected in the PES sample design. Taking into account sample loss and non-response expectations, approximately 43,000 dwellings were expected to respond to the PES, including around 800 dwellings from 33 Discrete Communities in NSW, Qld, SA, WA and NT. This represented an increase of 20% on the size of the 2011 PES sample.

GENERAL POPULATION SAMPLE

3 The General Population sample was selected through a stratified, multi-stage cluster design from the private dwelling framework of the ABS Population Survey Master Sample. Areas are grouped into strata based on locality, population density, remoteness, projected population growth and socio-economic characteristics. Each private dwelling within the stratum has the same probability of selection.

4 A focused sampling strategy was introduced into the General Population sample in 2011 to improve estimates for the Aboriginal and Torres Strait Islander population. The strategy was reviewed in 2015, which showed that most of the improvement in Aboriginal and Torres Strait Islander estimates was achieved by increases to the overall sample size, rather than from the focused design. As such, the strategy was discontinued in 2016.

DISCRETE COMMUNITY SAMPLE

5 For the purposes of PES sampling, Discrete Communities were grouped into sets comprising main communities and their associated outstations. The selection of main communities was undertaken with probability proportional to the size of the set. The aim was to select as representative a sample as possible while also considering cost constraints, reasonable interviewer workloads and expected sample size.

6 The sample selection process within the Discrete Community required the interviewer to compile a list of all the dwellings within the community. From this, a sample was taken using systematic equal probability sampling.

7 A selection of outstations associated with each selected main community was also included in the sample. Each outstation had an equal chance of selection and, once selected, all dwellings within the outstation were enumerated. Occasionally, an outstation was not enumerated if it was difficult to access from the main community resulting in a small loss of coverage.

2016 RESPONSE RATES

8 The total number of fully responding dwellings across both sample components, after allowing for sample loss, was 42,463. This represented a response rate of 91.2% for the General Population sample and 92.7% for the Discrete Community sample. This is a lower than 2011, which achieved 94.0% and 96.4%, respectively.

	NSW	Vic.	Qld	SA	WA	Tas.	NT	ACT	Aust.
GENERAL POPULATION
Households approached (after sample loss)	10 267	8 802	8 263	4 920	5 587	2 779	3 308	1 698	45 624
Households in sample	9 147	8 034	7 634	4 587	5 211	2 638	2 791	1 555	41 597
Response rate (%)	89.1	91.3	92.4	93.2	93.3	94.9	84.4	91.6	91.2
Persons in sample	24 713	21 153	20 434	11 519	13 677	6 230	7 921	4 177	109 824
DISCRETE COMMUNITY
Households approached (after sample loss)	28	–	124	65	153	–	564	–	934
Households in sample	22	–	115	65	137	–	527	–	866
Response rate (%)	78.6	–	92.7	100.0	89.5	–	93.4	–	92.7
Persons in sample	52	–	465	313	578	–	2 811	–	4 219