|Page tools: Print Page Print All|
SAMPLE DESIGN AND SELECTION
To achieve these design objectives, state and territory sampling fractions were set as shown in the following table, which also depicts the corresponding expected number of fully responding households. The sample selection procedures described below result in every dwelling in the same state or territory having a known probability of selection, equal to the state or territory sample fraction.
NATIONAL HEALTH SURVEY, State/territory sample
Actual numbers of fully responding households are available in the Response Rates sub-section.
Within each selected dwelling, one adult (18 years and over) and one child (0-17 years) (where applicable) were randomly selected for inclusion in the survey. This sub-sampling within households enabled more information to be collected from each respondent than would have been possible had all usual residents of selected dwellings been included in the survey.
Area-based selection ensures that all sections of the population living in private dwellings within the geographic scope of the survey (i.e. excludes very remote and Indigenous Communities) are represented by the sample. Each state and territory is divided into areas called strata. Each state is first split into capital city and rest of the state statistical areas, according to the Greater Capital City Statistical Area levels of the Australian Standard Geographical Classification (ASGC). These areas are then further split by a measure of socio-economic status as low, medium or high, based on proportions of households with different levels of equivalised income according to the 2011 Census. This forms the survey strata. In addition, each stratum contains associated 'area type' classes based on geographic classifications such as locality, population density, remoteness and projected population growth.
Each stratum contains a number of Base Frame Units (BFUs), the key area sampling unit. A BFU is typically a single Meshblock, the finest geographical unit in the ASGS, and will usually contain between 30-60 dwellings. These dwellings are split further into a number of clusters, with the number of clusters in a BFU based on the ratio of 2011 Census dwellings and a chosen cluster size, usually around 5 - 20 dwellings.
BFUs are first selected while controlling for socio-economic properties and geographical proximity. A cluster of dwellings is finally selected by applying a systematic skip to dwellings within the BFU.
Clusters of dwellings are selected with equal probability within strata.
In the 2014-15 NHS a total sample of approximately 21,850 households was selected which, taking account of an expected rate of sample loss (e.g. households selected in the survey which had no residents in scope of the survey, vacant or derelict buildings, buildings under construction, etc.) of 16% and an expected rate of non-response of 13%, was designed to achieve the desired sample of around 15,000 fully responding households.
To take account of possible seasonal effects on health characteristics, the sample was spread randomly across the 12-month enumeration period. Analysis of previous health surveys has shown no particular seasonal bias across key estimates. NHS was enumerated from 6 July 2014 to 4 July 2015 with the sample allocated randomly over four quarters, with a soft close (two week period) being implemented at the end of each quarter to allow for finalisation of any outstanding interviews:
** Q3 enumeration soft close extended until 11 April 2015
*** Q4 enumeration soft close extended until 4 July 2015
Sample Top Up
Sample top-ups are prepared during sample design and are used if it is found during enumeration that the responding unit count may be below survey requirements. The decision to use a sample top-up of 1,250 dwellings for Quarters 3 and 4 was made after there was greater than expected sample loss during the earlier stages of enumeration.
These documents will be presented in a new window.