6269.0 - Information Paper: Labour Force Survey Sample Design, May 2013  
ARCHIVED ISSUE Released at 11:30 AM (CANBERRA TIME) 30/05/2013   
   Page tools: Print Print Page Print all pages in this productPrint All



The LFS is designed primarily to produce reliable estimates at the national, then state and territory levels, but it also delivers estimates for a number of regions within states and territories.

LFS Statistical Regions were established to meet user interest in small area data from the LFS. These regions were originally established following analysis of data from Censuses of Population and Housing, extensive consultation with major users of labour force data, consideration of regional population levels required to yield reliable estimates, and the need for consistency with other statistical collections. Regional estimates are available monthly through Labour Force, Australia, Detailed - Electronic Delivery (cat. no. 6291.0.55.001). (Table 16).


The LFS Statistical Regions used for the publication of labour force statistics are based on the Australian Standard Geographical Classification (ASGC) (cat. no. 1216.0). The ABS has recently introduced a new standard, the Australian Statistical Geography Standard (ASGS) (cat. nos. 1270.0.55.001 to 1270.0.55.006) which supersedes the ASGC. The new sample design uses the ASGS mesh blocks. From August 2013, the whole LFS sample will be selected under ASGS. The data will continue to be published on the ASGC basis until the January 2014 issue, when the estimates are rebased using the 2011 Census of Population and Housing.

Until the completion of 2011 Census rebasing in the January 2014 issue, estimates will continue to be weighted and published under the current ASGC Capital City / Balance of State geographical boundaries and Labour Force Dissemination Regions. From the January 2014 issue onwards, Labour Force estimates and the regional time series will be published under the ASGS. These estimates will be re-weighted to the new ASGS Greater Capital City Statistical Areas (GCCSA) and Rest of State/Territory boundaries as well as the Statistical Area Level 4 (SA4) Labour Market Regions.

The time series will be reweighted and, depending on the outcome of further investigations, published back to July 1991 for the National, State/Territory, and GCCSA/Rest of State/Territory geographic levels and SA4 Level estimates are likely to be revised back to 2003.

As is the current practice, future sample designs are expected to see regional boundaries updated to align with the ASGS edition current for the Census on which the design is based. For further information see Australian Statistical Geography Standard (ASGS) (cat. nos. 1270.0.55.001 to 1270.0.55.006)


As with state and national estimates, regional labour force estimates are subject to sampling error. Compared with estimates at state level, estimates for regions are based on smaller samples and are subject to higher relative standard errors. The LFS is not designed to provide accurate regional estimates. The ABS has taken steps to improve the quality of the small area estimates from the LFS and the February 2004 survey introduced regional population benchmarks for estimating and releasing regional labour force estimates. These benchmarks are classified by LFS Statistical Region of usual residence and sex. The regional labour force estimates were revised back to January 1999 to include regional benchmarks and this change made a modest improvement in the quality of regional labour force estimates without compromising the quality of the national, state and territory estimates. However care should still be taken in the interpretation of regional estimates. The standard errors for each region are available through the product Labour Force Survey Standard Errors, Data Cube (cat. no. 6298.0.55.001). Overall, caution should be taken using regional statistics given the high degree of variability, particularly for the regions with smaller populations. Greater improvement in the quality of small area estimates from the LFS would require a substantial increase in sample size and would increase the cost of the survey and respondent load.

From the January 2014 issue, the number of labour force regional series will expand from 69 customised ASGC-based regions to the 106 standard ASGS SA4 Labour Market Regions. This finer level regional data will be subject to higher sampling error than ASGC dissemination regions due to the smaller populations in each region, but this will be offset against greater consistency and comparability between regional data across ABS products. Standard errors for each SA4 region will be available in a future edition of Labour Force Survey Standard Errors (cat. no. 6298.0.55.001) after the January 2014 issue release in February 2014.

For more information on the quality of estimates for the AGSC regions refer to the feature article Labour Force Survey Regions published in the July 2004 issue of Australian Labour Market Statistics (cat. no. 6105.0). This publication is available on the ABS web site <www.abs.gov.au> (Statistics).


The changes arising from the introduction of the ASGS, together with the introduction of the new sample (the new sample selected to represent each region may have different characteristics to the old sample) is likely to cause significant disturbance to Regional statistics from the LFS for May to August 2013 as the new LFS sample is gradually implemented. As a result of the method of implementation of the new LFS sample, regions in more remote areas will be more subject to disturbances than those in less remote areas. The ABS advises using annual average estimates for regions to minimise the variation over time in the sample.