Information needed Response
Title and link to statistical output Errors in official statistics publication – survey of childcare and early years providers
Name of producer organisation Department for Education
Name and contact details of person dealing with report Laura Selby
laura.selby@education.gov.uk
Name and contact details of Head of Profession for Statistics/Lead Official David Simpson
david.simpson@education.gov.uk
Link to published statement about the breach (if relevant)
Date of breach report 09/12/2021

Information needed Response
Relevant principle(s) and practice(s) Quality pillar
Date of occurrence of breach Issue with 2018/19 release, corrected 09/12/2021

The Survey of Childcare and Early Years Providers is a survey conducted by NatCen on behalf of DfE, and is an annual Official Statistics publication. In the process of analysing the 2021 survey data, NatCen discovered some serious errors in the weighting used for staff-level data, one of six main areas in the publication, which means that the data in 2018 and 2019 needs correcting.

To minimise potential confusion amongst users we are releasing the corrected 2019 figures alongside the release of the Childcare and early years providers survey: 2021 publication, scheduled for first release on 9th December 2021.

2018 release: https://www.gov.uk/government/statistics/childcare-and-early-years-providers-survey-2018
2019 release: https://www.gov.uk/government/statistics/childcare-and-early-years-providers-survey-2019
2021 release: https://www.gov.uk/government/statistics/childcare-and-early-years-providers-survey-2021

The weights affect the analysis relating to staff, which covers one of six chapters within the 2018 report and two of 25 tables in 2018 and 3 of 28 tables in 2019.

The change in weights have the biggest impact on the proportion of staff in school-based providers (SBP) qualified to Level 6. This has gone down from 48% in 2019 previously to 33% using the new weights. The proportion of SBP staff qualified to Level 6 is now similar 2018, 2019, 2021 and similar in all 3 years to the results from 2016.

There are more modest corrections (c.1-2 percentage points) to figures for group-based providers (GBPs) for:

  • Age profile of staff
  • Proportion of staff paid less than the national living wage
  • Qualifications of staff in GBPs.

As the published statistical releases use incorrect data, we will reissue the 2018 and 2019 reports on the same day as the new official statistics publication for 2021 is published (9 December). This is because the errors have only been confirmed in the process of finalising the 2021 publication, and the corrections will get a greater audience if they are published at the same time as the 2021 statistics.

For the 2018 report, both the main report and the tables will be updated but for 2019, only the tables will need to be updated as staff-level data was not included in the final report. The finance reports from 2018 and 2019 may also need to be corrected, although these are published as research reports rather than official statistics.

The 2021 report will include corrected 2018 and 2019 data as part of time series comparisons. The errors in previous reports will be flagged in the box outlining guidance on ‘Comparisons between 2021 and other publications’, with clear signposts to which sections were affected.

Change notes will be added to the 2018 and 2019 releases to explain the reason for change to users.
Natcen have thoroughly quality assured the 2021 weights and those used for 2018 and 2019, and have sent a detailed note of QA processes undertaken for 2021. Analysts within DfE have reviewed these processes and are content.