Findings from insights generated from data and their presentation
Understanding personal experiences
Ideas were discussed for gaining better insights from the data and engagement activities. These included approaches to better understand people’s life experiences, such as ethnography and other qualitative methods. Surveys were seen to predominantly focus on recording descriptive characteristics, providing little insight into the needs of individual.
“A lot of the data sources we’re working with are not based on the lived experience of disabled people but rather just simply outcomes, or management information, which usually are related to benefits or others and how people are given money, rather than what they need to survive.” (Central government participant)
“Less weight is given to perceptions and emotions, and basically, people’s perceptions of experiences, […] perceptions and feelings have great power, and we need to understand that to make better policy.” (Central government participant)
“What would a human-centred approach to data collection look like if it started with people and their life experiences, and how would that re-categorise the labels that we use to describe people and the topics and areas of interest that we even collect information on? What are the interests that are important to marginalised groups, that don’t sit on the other side of the room? What would they want to see studies on? I just really actually think it’s vital to policy development.” (Central government participant)
Understanding data and community needs
Learned society participants identified a need for government offices and departments to undertake more frequent consultations on key changes to the publication of official statistics, to gain insights into data user needs and ensure they are provided for.
“Some publications provide a lot of information on a certain vulnerable group or protected characteristic, and there needs to be a higher level of consultation or engagement before that is stopped or changed in some way.” (Learned society participant)
Local government participants highlighted issues that arise from making decisions based on an incomplete understanding of what is happening in their communities. This lack of insight was said to result in community needs not being effectively captured when developing policies, which can thereby exacerbate inequalities. Devolved administration participants highlighted that without better data enabling experiences between diverse groups to be distinguished, policymakers are not fully able to clearly define who is being left behind. Welsh Government participants also highlighted how important insights from available statistics on inequalities are not sufficiently driving impact, change and accountability conversations.
“That connection with the information we do have doesn’t always seem to happen. Because there’s some pretty stark statistics in there, even with the stuff we’ve got, so why is that not driving impact and change.” (Welsh Government participant)
Data presentation
Linked to the accessibility of data outputs, interview and roundtable participants discussed how the ways in which data are presented and explained can impact their interpretation and therefore potential to provide useful insight.
“Explaining that difference is a real difference, or at least it’s a quite strongly inferable one, that’s the point where we hit the challenge, because that’s the point where we are convincing audiences that what we are saying is credible, should be considered and isn’t just luck of the sampling bias, or luck of measurement.” (Central government participant)
It was agreed that social media had the potential to reach a wide and diverse audience. However, issues with using social media to present findings were discussed, including the risk of reports being misread, key data being distorted and used to communicate messages suited to the reader’s interest, or the full report not getting its due regard.
“Twitter in itself and that short byte size nature of it means that there is real distortion and potential for distortion of data.” (Northern Ireland Executive participant)
Academic participants presented one of the main reasons for data inaccessibility as a lack of mandatory provisions for data outputs within the research community and the means to ensure inclusivity of access to research findings.
“There is a real balance about trying to make the data accessible so it can be accessed by people who have visual impairment or disabilities but also that the information is easy to access for people who are not statisticians or researchers.” (Learned society participant)
Solutions to improve insight
To address the insight issues described, participants suggested using qualitative approaches to capture lived experiences and develop insights into what matters to communities to improve inclusivity. Participants suggested that the collection of qualitative data could inform policy work in a different way to statistics, help communities feel more included, and help address data disaggregation issues. However, it was stressed that qualitative data should be captured in a structured manner, and findings disseminated efficiently across different organisations and government departments to ensure groups are not repeatedly asked the same questions.
The creation of a centralised database for local government policy analysis was also recommended.
“[This could be] interrogated in a way that would give you multi-faceted ways of looking at it through different lenses.” (Central government participant)
Participants recommended improving the accessibility of data that are presented to the public through a variety of methods, including:
- clear bullet points
- lay language
- clear and simple tables
- maps and charts
- downloadable data in CSV format
- producing accessible websites, using social media platforms
- making the information as easily digestible as possible
It was also suggested that researchers need to consider people who speak English as a second language and others who may be less proficient in English when disseminating research findings. A learned society participant highlighted the importance of using data visualisation techniques to improve the clarity of key messages, “but also picking up the key points and limitations of the data.” In addition, ensuring access to easy-read versions or screen-reader compatible versions of statistical publications where required, and producing offline versions of certain publications where possible.
Conducting cognitive interviewing to understand how users comprehend survey data was suggested by a research funding organisation participant, as well as working with certain under-represented groups. Participants felt this would improve the way data and evidence are presented in the future, making insights more meaningful for diverse groups.
“[To test] different ways of showing the same evidence and seeing whether that resonates with how they best connect with it.” (Learned society participant)
“Funders could probably help each other to come to some standards around this kind of activity.” (Research funding organisation participant)
Certain participants highlighted the importance of being transparent about data gaps to help contextualise the disparities that are observed and make the findings less open to misinterpretation.
“Make visible the gaps, rather than just make visible the data?” (Welsh Government participant)