The Labour Force Survey: A General Description

In this section I briefly describe the use of mathematical statistics underlying the design and analysis of the Bureau's Labour Force Survey (LFS).

• The LFS provides Australia's official estimates of employment and unemployment.
• It is conducted throughout Australia every month.
• Each house and residential unit (incl. motel rooms, caravan parks, etc.) has a chance of selection.
• With a sample size of around 30,000 houses, this corresponds to some 65,000 persons (approximately half a per cent of the population of Australia) from whom responses are collected.
• The survey information is collected by a panel of around 800 interviewers.

The survey is subject to both `sampling error' (as it is not a full census), and also `non-sampling error', because measurement processes will always entail some error. Mathematics plays an important rôle in minimising the effect of these errors, whilst also keeping costs at an acceptable level.

Mathematics plays a further rôle in the use of the derived estimates. It helps the user to extract information from the data collected; in particular, extending the usefulness of the data by bringing-out patterns over time.