UPDATE ON RECOMMENDATION 7 FROM THE INDEPENDENT TECHNICAL REVIEW
The Executive Summary of an independent technical review into the Labour Force Survey (LFS) and the ABS response to the review's recommendations were released on the ABS website on 9 December 2014. For details see the November 2014 issue of Labour Force, Australia (cat. no. 6202.0). Recommendation 7 of the review related to the use of Bayesian techniques in identifying and resolving problems with the LFS. An update on progress with this recommendation was provided in the May 2015 issue, and this note provides a further update.
Recommendation 7 of the review and the ABS response are:
Recommendation 7: Bayesian techniques should be considered as a means of identifying and resolving potential problems in core LFS series.
ABS Response: Agree and is in progress (in relation to assessing the relevance of the techniques) with findings to be reported by June 2015.
While the ABS is investigating Bayesian and related techniques, the ABS will continue to use the current composite estimator methodology for the Labour Force Survey and considerable additional work is required before these techniques can be considered a viable alternative.
Following consultation with academics and stakeholders, the ABS has identified a number of indicator series and data sources that could potentially inform Labour Force estimates. This process considered a broad range of data sources that labour force analysts use to increase their understanding of the current labour market, including the data sources that have been assessed by the Department of Employment for their Leading Indicator of Employment.
These data sources are used by analysts in a variety of ways. In considering these different approaches, it is important to note that Bayesian techniques are one method of incorporating multiple sources of information, but there are also a range of other approaches. Recognising this, the ABS has now expanded its consideration of methods in which other information sources can be used, to more fully consider the possible multi-source methods.
Work is currently underway to further assess external sources of information and to determine how multi-source methods could be used to inform an estimate of current month labour force outcomes. For example, a Job Vacancies series provides some indication of labour market demand, but does not enable the direct estimation of the number of people employed or unemployed. Mathematical modelling can determine the relationship between the information source and the labour force outcomes, and the model can then be used to assess labour force outcomes from the information source.
No changes will be made to the estimation methodology for labour force statistics without extensive consultation, and if any change is proposed it will be communicated and explained well in advance of implementation.
The next update on this research is expected to be provided in the May 2016 issue.