Explaining how the data should be used
3. To enable users to more fully understand the uncertainty around the LFS data, ONS should be clearer in its communication around terms such as ‘volatility’ and ‘caution’. It should also link clearly to the data which demonstrates where this volatility and uncertainty exists to enable appropriate use of the data.
Actions taken and commitments made by ONS to meet the requirement
ONS has included more detail around how users should interpret LFS statistics in its bulletins, reference tables and in its supporting metadata, such as its quarterly performance and quality monitoring reports and its estimates of sampling variability published in Table A11 every month. For example, in its June release, ONS prominently flagged the differing trends across its different measures of employment and advised users to make longer-term comparisons. To help users to compare datasets, ONS published Table X12, providing a comparison of annual growth rates of Pay As You Earn Real Time Information (PAYE RTI), LFS and Workforce jobs (WFJ). ONS has said that it will continue to publish Table X12 while users find it helpful.
ONS also included more detail in its April 2024 Labour market transformation update explaining how the lower number of responses to the LFS in late 2023 led to increased volatility in the single-month and three-month movements in its labour market estimates. ONS has said that it will communicate to users any additional uncertainty for these or other high-profile LFS indicators that it publishes.
OSR judgement
ONS has provided more information to help users understand the uncertainty in the data and committed to continuing to do this. To further support users’ understanding of the uncertainty, ONS should be clearer on the sources of bias in the sample. It should also improve its signposting from all the publications that include statistics derived from LFS data to the tables that provide the accompanying data. For example, it would be helpful if the Employment in the UK release included links to information around variability and the volatility in the statistics which are found in Table A11 and Table X12.
We understand ONS’s position that because volatility is not isolated to age groups alone, it could be considered misleading if ONS were to just publish sampling variability by age group. However, we consider that being clearer about the key sources of sampling variability data, including age, is helpful for users. Commentary could be provided alongside any such clarifications to reduce the risk of misleading users.
Action remaining
ONS should consider how to further improve its information on the key sources of sampling variability in the data and ensure that information about uncertainty is well signposted from all its bulletins that use LFS-derived statistics.
4. To help users more fully understand how the reweighted data can be used at the sub-UK level, ONS should include more explanations about what the reweighting improvements mean for the use of this data.
Actions taken and commitments made by ONS to meet the requirement
In its April 2024 Labour market transformation update, ONS set out the considerations users should make when using different LFS and APS data and statistics, given the current range of weighting approaches being used.
OSR judgement
We consider this requirement to be met. ONS has set out the weighting that it has used for the following datasets: LFS person-quarterly, Annual Population Survey (APS), LFS household-quarterly and LFS two-quarter longitudinal. We also appreciate that ONS is listening to its users and exploring options for further reweighting.