Policies and procudures

The following sections detail Skills for Care’s policies and procedures relating to the Code of Practice for statistics.

1.  Release practice policy 

1.1.   Pre-announcement and timing 

The release date of Official Statistics will be agreed between Skills for Care and the Department of Health and Social Care (DHSC) each year. Annual publications will be pre-announced at least 6 months prior to publication. Any delay in the availability of Official Statistics publications will be announced via our Workforce Intelligence site on the home page. 

Official Statistics publications classed as Official or Accredited Official Statistics are issued at 09:30 hours on the pre-announced date. 

1.2.   Pre-release access 

The Pre-Release Access to Official Statistics Orders permit limited advanced sight of statistical releases in specific circumstances. Skills for Care limit this to the minimum number of people necessary to allow officials to:

• Respond to questions or make statements at the time of publication.
• Take action before or at the time of publication. 

Pre-release access is not the default. Access is allowed only in circumstances where the public benefit is likely to outweigh the possible detriment to public trust in official statistics and individuals satisfy the requirements laid out in the Pre-release Access to Official Statistics Order 2008. Access is given for no longer than 24 hours before publication. 

Skills for Care will publish the organisation name and job titles of everyone who has been granted pre-release access alongside each official statistics report.

 

2.  Revisions and corrections policy

 Scheduled revisions will be managed systematically, pre-announced and reflected in dissemination plans. Large changes to the scope or content of official statistics will include a period of consultation. 

Unscheduled revisions are those that do not fit into the managed pattern of revisions, for example errors or unforeseen changes in methodology. For unscheduled revisions we will:  

• Where results are affected by methodological changes, a description of the change will be given. 
• Explain corrections to statistics at the earliest opportunity and in a way that provides the most benefit to stakeholders. 
• Issue corrections in an orderly way, in a new version with the amended statistic. 
• Summarise the key impact, nature and extent of corrections. 
• Reserve the right not to publish some minor revisions if the resources required to issue them outweigh the benefits to stakeholders.    

 

3.  Data management policy 

3.1.   Data storage and security 

Adult Social Care Workforce Data Set (ASC‑WDS) data is stored securely within Amazon Web Services (AWS). The dataset includes limited personal data such as date of birth, National Insurance number, and local identifiers (which may occasionally include a person’s name). 

3.2.   Data de‑identification and processing 

Before any personal data becomes available to Skills for Care’s analysis team, it undergoes a formal de‑identification process in line with the principles of data minimisation, purpose limitation and security under the UK GDPR. This process includes:

Transforming dates of birth into an age variable, ensuring that the original date of birth is not retained.

Removing national Insurance numbers and replacing them with a unique pseudonymised identifier generated algorithmically.

Removing all direct identifiers, such as names and other local identifiers in their entirety. 

Following these processing steps, the data made available to our Workforce Intelligence Analysts are anonymised for analytical purposes, meaning that individuals cannot be directly identified from the datasets accessible to the team. 

In accordance with our confidentiality policy, we also apply statistical disclosure control measures during analysis and reporting. These additional safeguards further reduce the residual risk of re‑identification from aggregated outputs, supporting GDPR‑compliant handling of personal data throughout the process. 

3.3.   Access Controls 

Skills for Care’s analysis team stores and processes data within a private, access‑restricted area of Skills for Care’s secure IT environment. Access to this area is limited strictly to authorised members of the analysis team and is controlled through role‑based permissions, in line with the UK GDPR’s requirements for data security and access minimisation. These controls ensure that only staff with a legitimate business need can access the data. 

3.4.   Data Sharing 

Analytical datasets are not shared outside the Analysis team unless a Data Sharing Agreement (DSA) is in place. The DSA verifies that:

• The receiving organisation is a legal entity.
• Data will only be used for a legitimate and approved purpose.
• The receiving organisation agrees not to attempt to re‑identify individuals.
• Appropriate technical and organisational safeguards are in place to maintain the integrity and confidentiality of the anonymised dataset.

 

4.  Confidentiality protection 

This section describes the procedures used by Skills for Care to protect the confidentiality of individual workers and locations in all published ASC‑WDS outputs. It focuses specifically on statistical disclosure control (SDC) methods applied during analysis and publication. Security of the underlying data is covered separately in the Data management policy (Section 3 of this document). 

4.1.   Aggregation and row-level suppression

 Skills for Care never publishes row‑level ASC‑WDS data. No individual worker or location records are released publicly under any circumstances. All published statistics are aggregated and represent groups of workers, employers, or care settings. Direct identifiers or pseudonymised identifiers are never included in outputs or reports, ensuring that no individual or specific workplace can be identified from any published material. 

4.2.   Rounding rules

To reduce the risk of re‑identification through small numbers or unusual patterns, all counts in published tables and charts are rounded to the nearest standard unit. Rounding applies to:

• Worker counts.
• Workplace counts.
• Any totals or subtotals.

This prevents users from reverse‑engineering small groups or identifying individuals indirectly. A table of the rounding rules used by Skills for Care’s Workforce Intelligence team can be found on the Methodology page of the Workforce Intelligence website.

4.3.   Suppression of Small Bases 

Percentage breakdown suppression

Percentages are not published where the underlying base contains fewer than 5 workers.

This threshold prevents disclosure through unusually small denominators and avoids the risk that individual characteristics could be inferred in small groups.

Average (mean/median) suppression

Averages (for example of age and pay rates) are suppressed when calculated from fewer than 25 workers. This rule ensures:

• Averages are based on sufficiently large and stable samples.
• Outliers cannot disproportionately influence results.
• Disclosure risk is minimised.
• Published statistics remain robust and reliable.

These thresholds are applied consistently across all reports, dashboards, and external data releases.

 

5.  Public involvement and engagement strategy 

Official statistics serve the public good as public assets that provide insight, which

allows them to be used widely for informing understanding and shaping action. Skills for Care has several processes in place to ensure it gathers information from users in order to iterate and improve its datasets, reports and statistics.

Extensive user research is carried out during the continued development of the ASC-WDS. User research sessions take place during every new development and every change. This ensures that the service is user-friendly, adapting to changes in the sector and is a useful tool for employers.

Skills for Care invites feedback about all of its publications by providing by contact details on each report page on the Workforce Intelligence website. Contact information for the Analysis team is also given at the beginning and end of each written publication. By providing contact details in this way users are encouraged to comment and feedback about the reports and statistics they have used. These comments are regularly reviewed and taken into consideration for future editions.

The Workforce Intelligence website, which houses all of Skills for Care’s reports and data outputs, was re-developed in 2025. This re-development included user research sessions to ensure users could find and understand the information they need on our website. The website also includes a pop-up user feedback survey, asking users of the site if they were able to find everything they were looking for, how they rated the experience of using the website and offers the opportunity for users to provide additional feedback. This feedback is used to continually iterate the website and the data outputs housed there to ensure they continue to meet user needs

 

6.  Charges for additional services

Skills for Care publishes data on the reports and visualisations pages of its Workforce Intelligence website which are free at the point of use. Data is published in various formats (written reports, data visualisations and data download MS Excel files) and at various geographical levels to try maximise the user needs met by the public offer.

If a user has a need for bespoke analysis, or outputs that are not covered by our public offer, then a quote is provided based on the time it will take an analyst to produce this output. Skills for Care monitors these requests and if common themes occur, looks where possible, to add outputs to our public offer to meet these data needs.

The sharing of raw data files also carries an administrative charge to cover the time it takes a Workforce Intelligence Analyst to set up data sharing agreements, prepare the requested files and provide support on appropriate use of the files.

As well as bespoke analysis, Skills for Care can share raw data files for further analysis by external parties. The charge for sharing of these files includes an administrative fee to cover the time it takes a Workforce Intelligence Analyst to set up data sharing agreements, a fee to prepare the requested files and (where requested) a fee to provide support on appropriate use of the files and subsequent analysis.

For examples and further information on the types of bespoke analysis Skills for Care can provide, visit the Commission our Services page of our website.

Our approach to trustworthiness, quality and value

Quality

Quality: data and methods that produce assured statistics.

Almost all of Skills for Care’s Workforce Intelligence publications, including statistics for ‘The workforce employed by adult social services departments in England’ report, are based on workforce estimates created using data from the ASC-WDS. 

The ASC-WDS was designed for the collection and reporting of adult social care information. All definitions were created with, users and subject matter experts to ensure they are an accurate reflection of the sector and are clear and unambiguous. These definitions are reviewed regularly and updated where required. 

The Workforce Intelligence team uses the information collected in the ASC-WDS to create estimates for the size of the whole adult social care sector and characteristics of the workforce. Our methodology allows the analysis to represent all adult social care workers, even if the ASC-WDS has uneven levels of data coverage. 

Several methods are used to quality assure the data. Applicable data is validated at the point of entry into the ASC-WDS; data from large organisations and local authority employers is checked and approved. We use rigorous data quality checks which mean uncertainty and data limitations are minimised. Our methodologies are published for transparency and have been peer reviewed by universities and an independent statistician. If the methodology is improved from the previous year, then historical trend data is retrospectively updated to ensure any trends we show are ‘real’ and not the result of methodological changes. 

The Workforce Intelligence website has information about what we collect, our coverage, methodology, the integrity and data quality of ASC-WDS data and subsequent workforce estimates. Limitations of methods, data quality issues and inconsistencies with external datasets are explained to users at the point of data interpretation across all our reports regardless of format (written report, MS Excel, or data visualisations). We use plain English and have the users’ interest at the heart of all our published work. 

Throughout the year the team maintain and refresh our understanding of the use and potential use of the statistics and data, therefore keeping up with best practice and principles. We have lessons learnt after every publication to see where improvements to process or methodology can be made.