About Data About Us (case study)

Summary of the project

Never before has data played such an integral and granular role in how we live. On a daily basis, we are asked to make decisions about personal data about us – consenting to it being collected and used for many purposes.

We carried out a project with the RSA (Royal Society for the encouragement of Arts, Manufactures and Commerce) and Luminate to explore narratives around data rights and data ownership: how people feel about data about them, and how we can find ways to talk about data that everyone can understand and engage with.

Our aim was to discover what people understand and feel about data about them, and the rights and responsibilities they felt should exist around it. We held focus groups and a workshop, and developed a graphic and stories to guide the discussion. We produced a report and summary report, as well as a video and a podcast.

We believe people can’t own data, but do have rights over personal data about them. As our ODI Manifesto says: 'everyone must have the opportunity to understand how data can be and is being used' and 'everyone must be able to take part in making data work for us all'.

Key facts and figures

• Two focus groups took place in April 2019. Each was attended by 15 participants.
• One workshop took place in June 2019. It was attended by 13 participants.
• Since publication we have disseminated the work widely, at the MyData Conference in Helsinki, at Labour and Conservative Party Conference events, to a data team within the Department for Digital, Culture, Media and Sport (DCMS), the Office for National Statistics (ONS), the Centre for Data Ethics and Innovation (CDEI), the Royal Society, it has also been discussed on panels at MozFest, Digital Summit Dublin and at the ODI Summit. The work and our findings have been influential in opening up and moving forward the conversation about people’s understanding of data about them.
• Outputs included:
  • Report: About Data About Us
  • Summary report
  • #WeAreNotRobots video
  • #WeAreNotRobots podcast
• We also developed a graphic that explains the different types of data about us:
  • personal data
  • sensitive data
  • behavioural data
  • societal data.
• In the context of how data about them is used, people told us they want:
  • honesty and transparency
  • agency and control
  • compliance and enforceability
  • rights and responsibilities
  • context and fairness.

What was the ODI’s role?

We initially did desk research on data rights and ownership. This established ideas around what existing rights there are, and also what the commentary is around data ownership.

We created a graphic tool to help define the different types of data about us: personal, sensitive, behavioural and societal. Data about us comes in many different guises and can have a range of different impacts, not all always about the individual, but often about society as a whole.

Over the course of two focus groups and a workshop, we explored how members of the UK public feel about data about them. We asked what they felt about having ownership or rights around it, and what kind of control or protection they feel is missing or needs strengthening.

Based on the conversations and findings from the two focus groups, the workshop tested out the narratives about data. We decided to do this within a workshop environment so that people could work together in groups to construct stories and test their knowledge. The RSA led the workshops and we developed the stories and script.

When we tested the graphic with participants, we found that the conversations became more nuanced. It helped people to clearly and articulately tell us when they were happy to share and why, and what communication, transparency and engagement they’d expect from government and organisations about this.

People told us that they generally feel positive about the benefits brought by the internet and being more connected, but want greater honesty and transparency, agency and control, rights and responsibility, context and fairness, and compliance and enforceability over how data about them is used. Ultimately, they want to know that where data is concerned, they will be treated as people, not as robots.

The project produced five outputs. These included a report, ‘About Data About Us’; a printed summary report; a graphic that explains the different types of data about us; a short, shareable, animated video; and a podcast which brings to life some of the perspectives we heard and invites people to share their views with the hashtag #WeAreNotRobots.

We want to start a conversation between people, governments and businesses, along with NGOs, interest groups and think-tanks. We want to discuss the rights we have, and the responsibilities that governments, businesses and all of us should have. We want to strengthen our data rights.

We discussed our findings at the Labour party conference in the UK. We also presented the work at MyData in Helsinki, Finland.

What was challenging?

There was a set deadline of the Labour party conference. This meant timings were quite tight. We also had to deal with the political uncertainty at the time which caused inevitable difficulties in planning and communication.

The findings weren’t disseminated as widely within the political groups as we had wanted them to be. Our findings were set to be discussed at the Conservative party conference as well as at the Labour party conference, but the work was discussed as part of another panel, as opposed to one specifically hosted by the RSA.

For future projects, there should be a detailed dissemination plan so that we target those who are the most influential.

What went well?

The focus groups highlighted more knowledge of data protection than is presumed. We are often led to believe, by the press, parliament and in published surveys, that the UK public’s knowledge and understanding of data protection is low. However, the discussions had between people in our focus groups suggested otherwise. This was extremely useful in being able to shift forward conversations with policymakers and politicians. In terms of those in power knowing that the public understand data more clearly and have a relationship with data about them that is complex, there is the opportunity to make more nuanced and appropriate data policy going forward.

The graphic tool helped make people aware of what kind of data about themselves they are being asked to share, what data they have to share and why, and what data they would prefer not to share. This helped people better understand what they are being asked to consent to and can assist with improving education. When we talk about data about us it is easy to imagine that all data is the same. But data can be an identifier; can be more sensitive than other types of data; can provide or create insights; or it can be used to make decisions that don’t just affect us as individuals but can affect and improve society as a whole. Learning to think about different powers of data is important. Our graphic tool, Types of Data About Us, is a starting point of that conversation.

The enthusiasm we have when talking about the project had an impact. Renate Samson was asked to brief a team at DCMS who were trying to build a similar sort of project. We found that the messages went down well with groups such as these. Part of this was Renate’s enthusiasm for the project, but mostly it was that we spoke about our findings in a different way. The emphasis was on what people had told us, not our assessment of their views. By giving the public the platform to speak we have been able to show that the public are far more literate on data issues – and the impact it has on their lives – than policymakers often realise. By giving them a voice, we have arguably been able to demonstrate this more clearly than previous quantitative work has done.

The project enabled us to shift the conversation. We found that there isn’t a one-size-fits-all answer to this, as different people have different approaches, and any one person will take a different approach in different situations. Encouraging policymakers to understand this can be powerful. While this requires further research with larger cohorts, for now, the findings from this project can potentially shift the way that those in power ask questions of the public or listen to and assess answers. We have more widely learned that asking people to tell us what they think in an open way – rather than posing questions where nuance can’t be demonstrated – is enabling a more insightful conversation to take place. We would encourage this approach to continue. It may not bring ‘solutions’ but it will help us all to see where further conversation or wider literacy is needed.

For future projects, it is important to consider how to present information so that it enables a shift in the conversation and so has an impact on policymakers. It is also important to ask open and far-reaching questions at the research stage to gain richer insights.

What have we learned?

Keep workshops focused and the agenda tight. Although the workshop produced some interesting discussions, they did tend to expand into subjects that were out of scope, for example workers’ rights, privacy and surveillance. Data is a broad topic so conversations can become unhelpfully far ranging and require effective facilitation to remain on topic.

Although small, informal dissemination of the messages can have an impact, we need to ensure we’re getting the messages to the influential policy people. Renate Samson presented to the ONS, DCMS, and the CDEI, and was also part of an RSA panel for MozFest. But we need to ensure we’re talking to politicians, which has been difficult this year (2019) because of the political uncertainty and the general election in December.

We need to ensure we speak to a geographically diverse and representative range of people. The workshop and focus groups were in greater London only, which potentially restricted the views we gathered.

Qualitative data can be very insightful. The report was about qualitative rather than quantitative data; about having the conversations and reporting on the discussions. Quantitative data in this subject area wouldn’t have been as useful, as it’s about people making choices and decisions based on context.

You can ask people difficult questions. When designing user research about data, it is wrong to assume that people aren’t aware of the issues around data. Asking the difficult questions often leads to some really interesting answers.