Micro-funding for researchers in developing countries could help tackle everyday obstacles, say Brian Rappert and his colleagues.
The expectation, and indeed requirement, that publicly funded scientists make their data openly available is entering mainstream science policy in the UK and Europe. Other regions are following suit, aiming to enhance discovery and reproducability.
In 2016, the African Open Science Platform was launched to address issues of access and coordination. Many national science agencies in Africa have endorsed this commitment to open data, leading to a rising number of policies on data sharing and open access.
Alongside this high-level support, however, there is also widespread recognition of the challenges associated with increased openness and the importance of encouraging open science among researchers in less resourced environments.
Little is known about how scientists in such settings feel about sharing their data. In a recently published study we, along with our colleague Ann Kelly of King’s College London, have broadened the picture by investigating attitudes toward data sharing in four biochemistry labs in South Africa and Kenya.
By many measures, these were successful labs. They were publishing, securing international funding and collaborating with external organisations. Nevertheless, data sharing was uniformly rare.
Several weeks spent at each lab speaking with staff and postgraduate students made it clear that researchers were struggling to do their jobs. Not surprisingly, everyone we spoke to mentioned funding. This, however, went beyond a simple lack of money.
Interviewees described their difficulties in spending the money they had, owing to institutional regulations and funder constraints. This left researchers struggling to meet the everyday demands of expenses not covered by grant funding or core institutional support.
These include facility repairs, equipment maintenance and calibration, software, off-campus internet access and professional subscriptions. General challenges—such as power outages and limited data downloads—compounded these hindrances.
These labs are hobbled by a lack of flexible funds to tackle the mundane challenges that eat up their time and so restrict their ability to engage with data. While participants talked openly about these challenges in their local settings, many shied away from raising them with international colleagues and funders for fear of losing credibility.
This has led to an impasse. These scientists are unable to take advantage of the growing availability of online data. And the daily demands of research, which are unknown in richer settings, discourage data sharing. Researchers saw the risks of getting scooped or overtaken by better resourced competitor labs as far outweighing any perceived benefits of data sharing.
The result is data poverty within data riches. What can be done? Appeals to the universal values of science or the formal requirements of research funding cut little ice in low-resourced environments. A potentially more effective approach would be to link expectations of greater openness with help to enable research.
This shift of focus from enhancing openness to enabling productive use of data is important. Pragmatically, it makes visible the underlying causes of data poverty. Politically, it reframes the situation of scientists working in low-resourced settings. These researchers are prevented from sharing data by challenging working circumstances and perceptions, not professional deficiencies or low productivity. These conditions can and should be improved.
One way would be to provide relatively small amounts of flexible funds to enable researchers to address their most pressing challenges. A small sum—often less than £100—is enough to fix many problems. Small investments that target a range of day-to-day constraints, as defined by the researchers experiencing them, could vastly improve productivity and engagement with data.
Such micro-funding is analogous to the cash transfers and microcredit systems that have become common in development programmes. These have shown the potential to empower people by supplying funds that recipients themselves decide how to spend.
These programmes have devised a number of systems of regulation and evaluation. Micro-funding in science would require similar checks and balances; what is required now are pilot programmes to evaluate such measures.
Micro-funding will not solve the broad systemic inequalities in global science. But it could help resource-poor researchers begin to experience the benefits of data sharing.
Brian Rappert and Sabina Leonelli are professors in the sociology, philosophy and anthropology department at the University of Exeter. Louise Bezuidenhout is a research fellow in the Institute for Science, Innovation and Society at the University of Oxford. See also Data Science Journal vol 16, p44 (2017).
More to say? Email comment@ResearchResearch.com
This article also appeared in Research Fortnight