Compliance
Last updated on Tuesday, 9 July 2024 at 09:20am
Statement of voluntary application of the Code of Practice for Statistics
The UKHSA data dashboard publishes up-to-date statistics about health threats in England, including coronavirus (COVID-19).
There are different types of data displayed on the dashboard:
- Data that is published in reports classed as official statistics
- Data that is published in reports classed as experimental statistics
- Data that has not previously been published and is classed as management information.
Because different classifications of data are displayed, it is not possible for the dashboard to be classed as official statistics. It is also not possible to meet some other requirements of official statistics. For example, we cannot:
- Pre-announce the release time of data (usually 4pm every Thursday) via the release calendar
- Restrict early access to the data to a small number of people, although access to the weekly update in its final form is limited to the dashboard team
We have therefore voluntarily applied the Code of Practice for Statistics as follows:
Trustworthiness
The UKHSA data dashboard is produced by the UK Health Security Agency (UKHSA), an executive agency of the Department of Health and Social Care (DHSC).
These statistics contribute to the government’s response to the health threats. In the interests of openness and transparency, we want to make this data available to everyone.
The dashboard has been developed by a team of analysts and data scientists based at UKHSA, under the guidance of the Head of Profession for Statistics. The dashboard has a steering group including members from NHS England, the DHSC and devolved administrations. The steering group feed into discussions about which statistics we include on the dashboard.
Data are updated weekly. We know timely information is important to our users, so we commit to update the data at 4pm every Thursday. On the rare occasions when delays are unavoidable, we announce this on our homepage and explain the reasons.
We announce any unexpected changes to the data we release on the homepage and add an explanation to our What’s new or About pages. For example:
- Delays to data being released
- Changes to past data, for example, because positive tests for a respiratory virus have been reported in error and later need to be removed
- Changes to the way data are calculated, for example, to improve the accuracy of data collection
We try to limit these changes, but sometimes they are necessary.
Quality
Data are compiled from several appropriate sources:
- UK Health Security Agency
- Department of Health and Social Care
- NHS England
- Office for National Statistics
We maintain close contact with these organisations to make sure that our work is aligned. Representatives from a number of these organisations sit on the dashboard steering group.
The UKHSA data dashboard team has 5 sub-teams, who are responsible for maintaining quality in different areas of work:
- software development team (software developers, front-end developers) – ensure the visualisation of data is displayed correctly and on time, provides technical support and ensures compliance with regulations for data protection and accessibility. The team also looks after the code, repositories, ETL service, Application Programming Interfaces (APIs), service infrastructure, database, analytics, logs, resilience, consistency, performance and security.
- data pipeline team (data engineers) – receives COVID-19 data from various different sources and transforms the different datasets into calculations that feed into the dashboard
- data analysis team (data analysts) – leads the development of the metrics, data and metadata that are fed into the data pipeline; updates the data, quality assures it and draws up reports for relevant stakeholders
- user research and design team (user researchers, interaction designers, content designers) – consults users of the dashboard to identify their needs and translates these into content and design improvements and additions
- product management team (service owner, product owner, delivery manager, business analyst) – makes sure the dashboard product meets the requirements of users and stakeholders; manages the programmes of work of the other teams
Under the leadership of the Head of Profession for Statistics, our team of professional data scientists and analysts ensures the dashboard uses correct and appropriate data, processing methods, quality assurance and visualisations. Statistics, data and metadata are compiled using recognised standards, classifications and definitions. These are set out in the About page.
Documentation which explains the rationale, definition, methodology and caveats for each metric is provided in the metrics documentation section.
Value
We make sure the information remains relevant by inviting feedback from our users and conducting regular research with them. We have carried out over 100 individual user research sessions. We have a monitored feedback mechanism and conduct regular surveys and interviews. In response to user feedback, we have introduced, for example:
- implementing more accessible ways of displaying data on the dashboard such as using a tabbed layout
- allowing users to download data from every graph displayed on the dashboard
We test statistics with our users to make sure they are not confusing or misleading and make improvements as necessary. As the public are the main users of the dashboard, we test all visualisations with them to make sure that the messages they present are clear and understandable.
We present data in several ways, to suit different users and help aid understanding. This means the data is of value both to people looking at it for personal interest and to those using it for professional reasons. We provide:
- Visualisations, including graphs and arrows to indicate trends
- Simple headline figures
- Full data tables
- Data downloads in different formats (CSV, JSON)
- Application Programming Interfaces (APIs)
We have conducted an accessibility audit of the site and provide an accessibility statement. We are always trying to improve the accessibility of our data.