Regulatory guidance: Dashboard questions

Thinking about Trustworthiness, Quality and Value in the development and use of dashboards.

We are seeing an increased use of dashboards in the dissemination of official statistics and data across the Government Statistical Service. In this guide, we provide some questions that analysts producing statistics can use in applying the standards of the Code of Practice for Statistics – Trustworthiness, Quality and Value – when developing and using a dashboard.

These questions target the most prominent issues and challenges with dashboards, as a form of statistics dissemination, that we collated from our producer engagement. Producers should use these questions to challenge themselves and ensure that any dashboards they produce follow the highest possible standards.

Trustworthiness

Trustworthiness is about the people, systems and processes used for producing the dashboard.

Consider:

• Who decides what data are included in the dashboard? Are decisions impartial and free from political or organisational conflicts of interest?
• When will the data first be published and next be updated, and will new data or visualisations be added? Do you have a process in place to notify users of any updates to the dashboard?
• How do you intend to manage errors?
• How are you considering data sensitivities? Where there are multiple, granular visualisations, how are you ensuring that the confidentiality of data is protected?
• What is the status of the statistics? Are they official statistics, accredited official statistics, official statistics in development, or something else? Is it made clear on the dashboard what the status of the statistics is?
• Do you have sufficient resources and skills to be able to build and maintain a dashboard? Do you have plans in place to maintain the functionality of the dashboard and its software throughout its lifespan?
• Have you considered how you will archive the data contained within the dashboard when you decide to retire it?
• What software packages are available and suitable to host the dashboard?
• How can users get in touch if they have any questions?

Quality

Quality is about the data sources and methods used and communicating any limitations of them.

Consider:

• What are the data sources underpinning the dashboard? Have you ensured that relevant links are available for users to access other relevant statistical products and supporting information?
• How have you communicated the uncertainty around the statistics? Our published report on the approaches to presenting uncertainty in the statistical system can give some useful pointers on how to do this.
• How have you assured the quality of the data? If using data from administrative sources, our Administrative Data Quality Assurance Toolkit may be helpful.  Have you provided links to all the relevant quality information on your dashboard?
• What can you tell users about any potential limitations with the data? Is there a risk of misinterpretation? Make sure that users see information that is critical to support their confident use of the statistics.

Value

Value is about ensuring the data meets user needs.

Consider:

• Why has the dashboard been created? Is it user-driven?
• What are the questions that the dashboard seeks to answer?  How have you engaged with users to understand these questions?  (The GSS guidance has some questions producers can ask users to better understand their need.)
• Is a dashboard the right tool to convey the statistical messages and to provide users with all the relevant information?
• How accessible are the data in the dashboard?  Will you be able to meet the government accessibility requirements with the dashboard?
• If the dashboard is not fully accessible, have you considered how the data contained in the dashboard will be equally available to all? Are there other statistical products planned, e.g. bulletins or data tables, as part of a broader communication package, to achieve this? If there are other statistical products, are these linked to from the dashboard?
• Does the dashboard include or signpost to all the relevant data and information?
• Can users easily download and use the data to draw further insights?
• How do you plan to engage with users to ensure that the dashboard meets their needs as they change over time?