6602.0 - Microdata: Longitudinal Labour Force, Australia, 2008-10 Quality Declaration
ARCHIVED ISSUE Released at 11:30 AM (CANBERRA TIME) 10/12/2012 First Issue
|Page tools: Print Page Print All|
In order to calculate estimates relating to the Australian population, rather than the numbers of people in the sample, weights need to be used for each person to account for their chance of selection in the Labour Force Survey (LFS) sample. In other words, if someone has a 1 in 100 chance of selection, then their responses are taken as representative of the 99 other people who were not selected, so we multiply this person's data by 100 to account for the unsampled data. This multiplier is known as the weight.
Households selected in different states and territories have different chances of selection to account for the differing population sizes and dispersion (for more information refer to Information Paper: Labour Force Survey Sample Design, Nov 2007, cat. no. 6269.0). The people interviewed in these households also have different sex and age characteristics, and the sample may not capture an exact representation of people in Australia (too many or too few 15 year old males, for example). By applying appropriately sized weights for each respondent, estimates can be calculated so as to account for the bias in the different response rates across population groups, as well as the household's chance of selection.
The Weight for the current month [WEIGHT] variable in this release of the Longitudinal Labour Force Survey (LLFS) CURF are the same as the weights used to produce the estimates in the November 2012 editions of the main Labour Force, Australia publications (cat. no. 6202.0, 6291.0.55.001 and 6291.0.55.003). These weights relate to the Person level (weights are not provided for the Family or Houeshold/Dwelling levels as most data items relate to the Person level). They are generated using composite estimation, and cross-sectional (point-in-time) estimates produced from the CURF will closely match the published original estimates (for more information, refer to Estimation method).
However, the weights have no relationship to the estimates published in the supplementary survey publications, as these are calculated using different benchmarks, estimation methods, and population exclusions. Weights from the supplementary surveys are not provided in the CURF. When calculating estimates based on supplementary survey data items in this CURF, they may not match official estimates produced in the publications corresponding to those supplementary surveys.
For a respondent who is in the labour force survey for the full 8 months cycle, they will have a corresponding weight for each month. These weights are then useful for calculating cross-sectional estimates, as they correspond to each 'point in time.'
For longitudinal analysis, there are a few different ways of calculating estimates based on the weights provided. These include:
Labour Force, Australia: Explanatory Notes - Population Benchmarks and Estimation Method (cat. no. 6202.0)
Labour Statistics: Concepts, Sources and Methods: 17. Overview of Survey Methods - 17.57, 17.64 to 17.66 (cat. no. 6102.0.55.001)
Labour Statistics: Concepts, Sources and Methods: 18. Methods Used in ABS Household Surveys - 18.42 to 18.43 (cat. no. 6102.0.55.001)
Labour Statistics: Concepts, Sources and Methods: 20. Labour Force Survey - 20.19 Estimation (cat. no. 6102.0.55.001)
Labour Force Survey Standard Products and Data Item Guide: Civilian Population (cat. no. 6103.0)
Information Paper: Forthcoming Changes to Labour Force Statistics, 2007 (cat. no. 6292.0)
Labour Force, Australia: Understanding Labour Force (cat. no. 6202)
Australian Demographic Statistics (cat. no. 3101.0)