Dear Ed,
In light of the suspension of the accredited official statistics status for labour market statistics from the Labour Force Survey (LFS) in February 2024, we have reviewed compliance with the Code of Practice for the Wales statistics derived from ONS’s Annual Population Survey (APS).
Due to the larger achieved sample on the APS for Wales, we understand these estimates are not affected to the same extent by the quality issues experienced by the LFS. However, the APS sample size has fallen and the statistical uncertainty for estimates has increased. I believe that many of the national level estimates continue to provide a reasonable indication of trends for Wales, particularly if used as part of a basket of indicators, however estimates for smaller geographies or population sub-groups are less reliable and may no longer meet user needs.
Additionally, unlike the LFS, the APS has not been reweighted with more up to date population estimates and therefore it does not take account of the significant changes to the size and structure of the population as suggested by the 2021 Census. The APS weighting methodology uses an adjustment based on older Pay As You Earn Real Time Information (PAYE RTI) data. The further we move away from the PAYE RTI timepoint the more unreliable the APS estimates will become. The APS population totals will be inconsistent with those used on the LFS and the discrepancy between the LFS and APS estimates may also increase over time.
Given these factors, I am writing to ask you to consider a temporary suspension of the accredited official statistics status for the estimates for Wales produced from the APS. This would affect the following sets of statistics due to be published on 09 October 2024:
This approach is in line with our understanding of ONS’s position and plans for their next APS publication. A temporary suspension of the accreditation in Wales will therefore ensure a clear and coherent approach for users of all APS estimates.
We will continue to ensure we are transparent with users on how reliable the APS estimates are. We will continue to provide appropriate advice and caveats in our publications for users of APS data and will apply suppression where appropriate.
If you agree to this suspension, this will mean that both the key sources of labour market estimates for Wales will no longer be considered accredited official statistics. Given that this will pose challenges for users, we will continue to work with ONS on the communication of labour market estimates for Wales to ensure they are used appropriately whilst a suspension of accreditation is in place. We also plan to ask ONS to undertake some analysis to help inform further understanding around the APS, particularly regarding the representativeness of the sample for Wales. We will, as always, keep users and yourselves abreast of any further developments.
Yours sincerely,
Stephanie Howarth
Prif Ystadegydd, Llywodraeth Cymru
Chief Statistician, Welsh Government
Related Links:
Michael Keoghan to Siobhan Tuohy-Smith: Request to suspend APS accreditation