While there seems to be general agreement that a broad field of research about data is emerging, it appears this field is still highly fragmented and this makes it difficult to conduct the type of multidisciplinary work that is often required in this space.
As discussed in the previous section, the specific focus areas of many funders cover part, though not all, of what we believe to be the landscape of data-related research. If this is the case, then this begs the question of why we feel it is important to consider all of these various areas of work as part of one larger data-related landscape. At the ODI, we believe that there are so many overlapping interests and areas of knowledge and expertise within the landscape that there is value in bringing the disparate parts together. Because data is important in so many social, economic and political contexts, a lot of data-related research is – or at least probably should be – multidisciplinary. Conducting research in one part of the landscape often requires bringing in knowledge and expertise from other parts. As long as the field remains dispersed and somewhat siloed, however, it will be difficult to acquire relevant knowledge or gather the necessary skills and expertise within one consortium, organisation or team.
The intersection between data, artificial intelligence and society, for instance, is one area of the data-related landscape that often requires multidisciplinary skills and expertise. According to researchers at Social Innovation Exchange: ‘Foundations all over the world are grappling with their role in the emerging field of data and artificial intelligence’ and yet, according to them, ‘few big foundations have the capacity or technical knowledge to either shape innovations or make sense of which ones to back, and when they do get involved they face complex challenges about transparency, ownership, and ethics’.
An interviewee from a UKRI research council made a similar point about efforts to increase data sharing in the health sector:
“It’s not just supporting one specific [type of project]. This is a huge area and involves multiple expertise, from statistics, computational science, mathematics, software engineering, data engineering, data stewardship, I mean, it’s vast, […] it’s a lot of different skills that you need to support. It’s hugely multidisciplinary […]. So on top of the discipline skills that you have to have, you have to support a variety of other skills. This is an area [where] I think […] we’re seeing more funding coming in [and] I think that we’ll see even more in the future as well.” (Participant G, UK research council).
This interviewee’s point about an increase in funding for multidisciplinary projects related to the sharing of health data is worth exploring. They seem to be suggesting that in the future there is likely to be increased funding for the type of multidisciplinary work that is needed for much data-related research and development - at least in the health space. Some of our interviewees, however, felt that multidisciplinary projects can be challenging to fund, with some funders prioritising more established, defined disciplines and research projects that sit in one section of the data-related landscape rather than straddling numerous topics or disciplines. According to one person from a philanthropic funder:
“[Multidisciplinarity] is vital to how the research works, how [researchers] think about things in these multidisciplinary ways. It was definitely a push in public funding to work in multidisciplinary ways. I probably don’t see that anymore from public sector funding, or as much as it was maybe five years ago. But I think universities are still trying to think like that” (Participant C, philanthropic funder).
The relationship between multidisciplinarity and funding therefore seems somewhat uncertain, with some people thinking that funding for research sitting across different fields will increase, and some thinking there has already been a decline. The impact this will have on cross-domain data-related research is also uncertain at this time.
To assess whether or not there has been an increase or a decline in data-related research funding as a whole, we conducted a quantitative analysis of the funding portfolios of two governmental funding organisations that we interviewed. Our hope was to identify whether the amount of funding they directed toward data-related research over the past decade had increased or decreased. However, as the quantitative findings section will illustrate, it has proven difficult to produce a definitive answer to this question within the scope of this work, and further research will be needed in order to investigate this question in more depth.