Dear Darren
ONS Labour Disputes statistics
We recently completed our compliance check of your labour disputes statistics against the Code of Practice for Statistics. These statistics cover three main measures of labour disputes; number of working days lost, number of stoppages of work and the number of workers involved in strike action from UK businesses. With continued public debate around employee terms and conditions and increasing living costs, these statistics are a valuable data source to support wider public understanding of a very current and relevant issue. It shows excellent responsiveness from ONS that following the temporary suspension during the COVID-19 pandemic, these statistics were reinstated in 2022. Data collection using the Labour Disputes Inquiry restarted in June 2022 and it was good to see that ONS published an article in September summarising its progress and future publication plans in an open and transparent way. Since November 2022, data are now included and published routinely as part of the monthly labour market release, six weeks after the reference period.
Overall, our review found that the statistics comply with the Code of Practice and should continue to be designated as National Statistics. This letter presents our findings, including feedback to address as you continue to develop these statistics. Our review found several positive features that demonstrate the trustworthiness, quality and value of the statistics:
- Users can access the data via the workplace disputes and working conditions ONS webpage and as part of the monthly UK labour market release supporting a coherent and joined up narrative of the UK’s wider labour market. Similar surveys are carried out in all Organisation for Economic Co-operation and Development (OECD) countries. ONS provides the International Labour Organisation (ILO) with annual data for publication on its webpage to further support reuse.
- The number of working days lost is the preferred indicator, as noted in the data table and the estimates are rounded in the monthly labour market bulletin; we consider this a good approach recognising a degree of uncertainty in the data.
- The ONS team is rebuilding a network of contacts with key users such as the Department for Business and Trade, the Advisory, Conciliation and Arbitration Service (ACAS) and trade unions to inform the ongoing development of the statistics. This network will also help to promote the survey to businesses and the importance of completion. This broader engagement with users should help harness insight gathered to help inform and direct ONS’s future publication plans, and should helpfully supplement feedback received via the ONS labour market inbox.
We also identified areas that would enhance the quality and value of these statistics:
- The survey is different from most other business surveys conducted by ONS in that it is voluntary and is a ‘census’ of all organisations identified as being involved in labour disputes within a particular month. The businesses are identified based on any strike action picked up by ONS through media coverage and information taken from ONS’s extensive trawling of trade union websites and social media. Given the relatively unusual nature of establishing the sample for these statistics, we consider that you should publish explicit information (such as by updating the Quality and Methodology Information (QMI)) about coverage, and the impact that this may have on the quality and level of uncertainty of the estimates.
- Where possible, the data are collected directly from the employer. In most cases, the survey questionnaire is in Microsoft Word format and circulated via email to businesses, which on occasion, can be misinterpreted as spam or illegitimate contact and lead to non-completion. It is good to hear that the team is seeking ways to overcome this and is hoping to align the collection of data with other ONS business surveys using the Survey Data Collection (SDC) system.
- The ONS team told us that survey response rates are declining. The Labour Disputes Inquiry QMI notes that the average response rate for 2015 was 84%, yet this is out of date, with latest response rates for December 2022 much lower. Published information about response rates should be kept up to date.
- Data not collected directly from businesses are imputed based on information gathered from previous returns and trade union or media websites. Methodology was last reviewed as part of the Triennial Mini Review of the Labour Disputes Inquiry in 2009. We recognise the bespoke nature of the imputation method used in this survey, but consider that the method and any supporting information should be reviewed as part of the ongoing development of these statistics.
I would like to thank the members of your team for their positive engagement on this review. We recognise that there are resource constraints within your team, however given the current relevance and public value that these statistics provide, we would ask that you consider ways to prioritise developments. Our Labour Market and Welfare domain lead will continue to engage with you and your team on progress in the coming months. Please do not hesitate to get in touch if you would like to discuss any aspects of this letter further.
Yours sincerely
Mark Pont
Assessment Programme Lead