Australian Bureau of Statistics
Celebrating the International Year of Statistics 2013

USE OF WEIGHTS

On each level of the CURF, every record contains a 'weight'. The weight indicates how many population units are represented by the sample unit. The CURFs contain records which can be adjusted (weighted) to infer results for the total in-scope population in Australia. As the survey was conducted on a sample of private dwellings in Australia, it is important to take account of the method of sample selection when deriving estimates from the CURFs. If the chance of selection is not accounted for, by use of appropriate weights, the results will be biased.

Each household and person record contains a weight. The identifiers are:

• Person weight: MHSFINWT, and
• Household weight: HHFINWT.

The household weight can be found on the household level, while the person weight can be found on the person level. This weight indicates how many population units are represented by the sample unit.

Where estimates are derived from the CURF, it is essential that they are calculated using the correct weight for that level of the file, and not just counting the number of records in each category. If person or household weights were to be ignored when analysing the data to draw inferences about the population, then no account would be taken of a person’s or household's chance of selection, or of different response rates across population groups, and the resulting estimates may be seriously biased. The application of weights ensures that estimates conform to an independently estimated distribution of the population by age and other characteristics, rather than to the distributions within the sample itself.

It should be noted that as a result of some of the changes made to protect confidentiality on the CURF, estimates of benchmarked items produced from the CURF may not equal benchmarked values. For more information about these differences see 'About the Microdata'.


Replicate weights

Replicate weights have been included on the CURFs to enable the calculation of sampling error for each estimate produced. Sampling error arises because the estimates are based on a sample of units and so will differ from estimates that would have been produced if all units in the population had been included in the survey. The CURFs contain:

• 60 person replicate weights (WPM0101 to WPM0160); and
• 60 household replicate weights (WHM0101 to WHM0160).

For more information on sampling error see 'Reliability of estimates'.

Previous Page

Next Page