WEIGHTING AND ESTIMATION PROCEDURES
WEIGHTING
A sample survey is the process of collecting statistical data from a sample of the in-scope population. The in-scope population for PSS 12 was defined as all persons aged 18 years and over living in private dwellings across Australia (refer to Scope of the survey). To select the sample of persons for the 2012 PSS, a sample of households was first selected. Then, from each household, one-in scope person per household was selected.
Weighting is the process of adjusting results from a sample survey to infer results for the total in-scope population. To do this, a 'weight' is allocated to each person that responds to the survey. The weight is a value which indicates how many people in the in-scope population are represented by the responding person. The use of weights ensures that estimates will conform to an independently estimated distribution of the population by certain categories including state, age and sex rather than to the distributions within the sample itself.
The steps used to derive person weights are described below.
Selection weights
The first step in calculating weights for each person was to assign an initial weight, which was equal to the inverse of the probability of being selected in the survey. The initial person weight was derived from the initial household weight (inverse of the probability of the household being selected in the survey) multiplied by the total number of in-scope males or females in the household depending on the assigned gender for the household. For example, if the probability of a household being selected in the survey was 1 in 600, and the household contained 3 in-scope females, then the selected person within the household would have an initial weight of 1,800 (ie 600 x 3). That is, the selected person in the sample represents 1,800 other persons in the population.
Benchmarking and non-response adjustment
Given the relatively low response rate for the 2012 PSS, extensive analysis was undertaken to ascertain whether there were any particular categories of persons that were over or under-represented in the sample. This over or under-representation in the sample can be corrected using a non-response adjustment and/or through calibrating the weights to population benchmarks. Both of these approaches were adopted in PSS12 and are described below.
Non-response adjustments can be used where information is available for all dwellings (responding and non-responding). For PSS12 interviewers collected observations for all dwellings and investigations showed that there was a correlation between the interviewer observations and whether or not a fully responding interview was obtained. This information was used to apply a non-response adjustment to the initial weights prior to calibration to benchmarks.
Benchmarks are independent estimates of the size of the population of interest. Weights are calibrated against independent population benchmarks to ensure that the survey estimates conform to the independently estimated distribution of the population, with respect to the benchmark categories, rather than to the distribution within the responding sample itself.
The PSS survey estimates were benchmarked to the estimated resident Australian population aged 18 years and over who were living in private dwellings (excluding very remote areas of Australia and those living in Discrete Aboriginal and Torres Strait Islander communities), simultaneously using the following benchmark categories:
Number of persons by -
- State or territory by capital city/balance of state by age groups by sex;
- State or territory by Social marital status (Married in registered or de facto marriage and Not married);
- State or territory by broad Country of birth (Australia, Main English Speaking categories and Other); and
- State or territory by Labour force status (Full Time Employed, Part Time Employed, Unemployed, or Not In the Labour Force).
The benchmark categories for State or territory by capital city/balance of state by age groups by sex were sourced from the Estimated Resident Population at mid point of survey enumeration (refer to Quarterly Estimates of the Australian Resident Population in
Australian Demographic Statistics, Jun 2012 (cat. no. 3101.0)). This Estimated Resident Population in-scope was estimated to be 17,201,700 as at 30 June 2012. The benchmarks, and hence estimates from the survey, do not (and are not intended to) match estimates for the total Australian resident population in
Australian Demographic Statistics, Jun 2012 (cat. no. 3101.0) (which
include persons living in non-private dwellings, such as hotels and boarding houses, persons living in very remote parts of Australia and those living in Discrete Aboriginal and Torres Strait Islander communities) obtained from other sources.
The remaining benchmark categories were based on pseudo-benchmarks. Benchmarks are considered to be 'pseudo-benchmarks' if they are obtained from a sample survey and as such, have a non-negligible level of sample error associated with them. The monthly Labour Force Survey (from February to December 2012) provided the pseudo-benchmarks for labour force status, social marital status and country of birth categories. The pseudo-benchmarks were aligned to the resident population aged 18 years or more, who were living in private dwellings in each state and territory, (excluding very remote areas of Australia and those living in Discrete Aboriginal and Torres Strait Islander communities), as at 30 June 2012. They were also made to represent the same population as the demographic benchmarks with respect to state/territory, part of state, age group and sex. The sample error associated with these pseudo-benchmarks was incorporated into the standard error estimation.
ESTIMATION
Information for sampled persons is multiplied by the weights to produce estimates for the whole population. Estimates of counts of persons are obtained by summing person weights with the characteristic of interest. All the estimates contained in this publication are based on benchmarked person weights.
If each person's weight were to be ignored when analysing the data to draw inferences about the population, then no account would be taken of each person's differing chance of selection or of different response rates across population groups, with the result that the estimates produced could be seriously biased. The application of weights ensures that estimates will conform to an independently estimated distribution of the population by certain categories including state, age and sex rather than to the distributions within the sample itself.
Users should take into consideration the quality of the estimates when interpreting data from the PSS. For further information about how to calculate RSE's and conduct significance testing, refer to
Data Quality. For other related information on other factors to consider when interpreting results, refer to
Interpretation of Results.
Print Page
Print All