Diversity, equity, equality, inclusion and accessibility Level: Foundation
Inclusive data practice
How the playbooks help you meet equality, diversity, inclusion and accessibility duties in your data work, and how this portal is designed so that everyone can use it.
In brief
- Poor or missing data often falls hardest on the people services most need to reach.
- Each capability has its own inclusion checks; use them at the steps where decisions are made.
- This portal is designed to be usable by everyone, and says where it falls short.
Duties that apply
- Public sector equality duty (Equality Act 2010)K: General knowledge, not checked against a source in this release
Have due regard to eliminating discrimination, advancing equality of opportunity and fostering good relations, across the protected characteristics. - Public Sector Bodies Accessibility Regulations 2018K: General knowledge, not checked against a source in this release
Websites and documents you publish, including data quality statements and catalogue pages, must meet accessibility requirements. - UK GDPR accuracy principleP: Practitioner material from cross-government playbooks not yet published
Personal data must be accurate and, where necessary, kept up to date. Inaccuracy can fall harder on some groups. - Welsh Language Standards and devolved dutiesK: General knowledge, not checked against a source in this release
Where your organisation is subject to them, statements and records meant for the public must meet them.
Duties marked K are general knowledge: confirm how they apply to your organisation with your equality and legal advisers.
Inclusion checks for each capability
Data governance
- Include equality considerations in the charter and the board's terms of reference
- Make sure owners and stewards have accessible training and materials
- Record decisions that affect groups differently, with the equality analysis behind them
- Invite representatives of people the data describes when the board sets priorities
Data cataloguing
- Write titles and descriptions in plain English and expand acronyms
- Record known gaps in coverage - people, places or periods the data misses
- Check address, name and identifier fields work for all four UK nations and overseas
- Make sure catalogue pages and records meet accessibility requirements
Data quality
- Break quality results down by group where you lawfully can
- Record poorer quality for a group as its own issue, so it is not averaged away
- Fix forms that force wrong answers, such as no 'prefer not to say' or one-line names
- Test quality statements with people outside the data team, and offer accessible formats
Questions to ask about any critical data asset
Questions considered0 of 7 done
- Does the data cover everyone it should? Compare its coverage with an independent count by group and place.
- Is missing or wrong data worse for some groups?
- Can people describe themselves accurately in the forms and systems that create the data?
- Do rules and records work for all four UK nations, and for Welsh language users where required?
- Are the outputs (records, statements, reports) accessible and in plain English?
- Have people the data describes, or their representatives, had a say?
- If the data feeds automated decisions or AI, has bias been assessed?
How this portal is designed for everyone
- Plain English, short sections, sentence case and a summary at the top of every page
- Works with a keyboard, screen readers and magnification; tested automatically against WCAG 2.2 AA
- Light and dark modes, larger text and wider line spacing, chosen in display options
- No information conveyed by colour alone; every chart and graph has a table equivalent
- Pages still work with scripts turned off; progress tracking is an optional extra
- Progress is kept only in your own browser; nothing about you is collected
Read the accessibility statement, or change text size, spacing and colours in Display options at the top of every page.
Was this page useful? Tell us Tell us privately, without GitHub