Findings

The following findings are derived from the feedback of participants in this research and aim to improve the understanding of what the ‘public good’ is from the perspective of members of the public. ADR UK and OSR will consider these findings, alongside other evidence, to inform how their work can maximise the benefits associated with data for research and statistics.

1. Public Involvement

Members of the public want to be involved in making decisions about whether public good is being served

Participants expressed a preference for meaningful public engagement to help inform decision-making concerning the use of data for research and statistics and explored several forms of how the public could be involved in decision-making. Participants articulated that inclusive public panels, with diverse members of the public, should play a central role in decisions made about data and statistics. They suggested that this role be supplemented with public conversations around the wider use of data for research and statistics, with requirement for continuous efforts to engage with the public.

2. Real-World Needs

Research and statistics should aim to address real-world needs, including those that may impact future generations and those that only impact a small number of people

Participants suggested that the value of data being used for research or statistics should be assessed by need, rather than by the number of people who would benefit, suggesting that serving the public good does not refer to serving the needs of a specific number of people. Participants felt that addressing social inequity and social inequality was a particularly pertinent reason to enable access to data for public good.

Participants asked that the public have full and transparent access to the decision-making process of Data Access Committee (see Appendix A Glossary of terms) to understand how public good was intended to be served. Although this information may already be publicly available, participants felt it could be more easily accessible. Participants articulated that they would like to see transparency from Data Access Committee regarding the impacts of proposed projects, including on how projects aim to address issues related to equality and inequity.

3. Clear Communication

To serve the public good, there should be proactive, clear, and accessible public-facing communication about the use of data and statistics (to better communicate how evidence informs decision-making)

Participants felt strongly that the public would benefit from greater awareness of the practices, motivations and outcomes of public good use of data for research and statistics. While information does exist on websites, and is shared across social media channels, participants felt these messages often did not penetrate their personal networks. Proactive communication that is clear and accessible – both regarding the use of language and availability of information, with the aim of reaching broader audiences, was viewed as a solution.

An example suggested was a national campaign to raise awareness about the public good use of administrative data for research and statistics. Public awareness of data use and associated practices was perceived as a way to support further democratic accountability for those who are responsible for ensuring data is used in a secure way.

4. Minimise Harm

Public good means data collected for research and statistics should minimise harm

Many participants felt a personal responsibility that data about them should not contribute to anything harmful; for example, data should not be used to perpetuate stereotypes about certain groups of people. To mitigate potential harms, participants suggested consulting members of the public, particularly those with lived experience, about potential uses of data for research or the interpretation of statistical patterns.  Engagement with those who have relevant lived experience may particularly inform appropriate interpretations of statistics, including language.

The Five Safes framework was explained as an example of data security measures, for which participants showed support. As well as wanting more widespread public awareness of the security around data access, participants desired increased accountability from those working with data and statistics, though with whom precisely was not explored. Suggestions given were having a named ‘data protection lead’, a whistle-blowing procedure if misuse of data was identified, a public telephone line, and public awareness of the repercussions of data misuse.

5. Best Practice Safeguarding

Universal application of best practice safeguarding principles to ensure secure access to data should help people feel confident to disclose data

For ‘good’ to be truly realised, participants felt that a framework such as the Five Safes should be universally applied for the public to feel confident that public sector data is being used in a way that they can trust. Further, participants felt that even more data collected by public services should be securely stored and linked, and good quality evidence should be shared, in order to inform policy and decision-making. Some participants expressed wanting to maximise the use of available data via more data linkages to better understand multifaceted needs.

Participants had sympathy for the variety of reasons some members of the public may have for not wanting to disclose their data to public services. However, even participants who would prefer not to disclose their data wanted decision-making to be based on evidence representing everyone in society. Participants suggested research was needed to understand why people do not want to disclose data about themselves, alongside greater awareness of the role of administrative data in research for the public good. Related to this were hypothetical discussions around synthetic data potentially filling in gaps where datasets were patchy or lacked enough information. There was consensus that this was not a public good use of data, though participants were supportive of the use of synthetic data for other things, such as training or developing code, as long as synthetic data wasn’t being used in place of real data to make decisions.