How the way scientists share data is changing

As a Better Science Through Better Data 2018 writing competition winner, I was invited to share my thoughts from #SciData18.

Like Comment

Last November, Springer Nature convened a conference on the role of communication and data in scientific research with dozens of researchers from around the world. The scientific method relies first on collaborations between researchers and the consensus that arises from peer review, and according to a Springer Nature survey of over 7,700 researchers, sharing data with a wider audience is a highly rated priority for 76% of respondents.

As disciplines rethink how best to share the results of science with a wider audience, a new paradigm in data sharing and interaction is beginning to emerge. According to Dr. Rebecca Boyles, a senior manager in bioinformatics at RTI International, a new focus on a data commons is emerging. Dr. Boyles urged creating an effective environment for data sharing by focusing on cultural change. The goal, she said, is to move from the “superficial interaction” of emailing data to an institutional environment that fosters deeper interactions with others. In person, interdisciplinary interaction can incentivize better science, but according to Magdalena Skipper, chief editor of Nature, another way to encourage interaction is to provide recognition for the individuals involved in data creation. Recognition via data repositories, publishing data papers and providing citations for use of datasets could benefit both specific researchers and ultimately the wider scientific community.

Two specific examples showed the benefits of data sharing. Dr. Natalia Tejedor Garavito, a research fellow at the University of Southampton, shared how the publication of open-access data papers through Scientific Data led to a better understanding of the lab’s research and increasing recognition as their demographic data was subsequently cited almost 100 times and used during the Zika outbreak.

Similarly, Professor Alasdair Rae of the University of Sheffield showed how the sharing of a single map on FigShare received over fifty thousand views and interactions with fellow researchers and the general public via social media. Rae shared how the geographies based on transport informed an epidemiologist’s work on disease by providing a new way to look at human movement. Throughout the day, different ways to share data effectively were presented. John Burn-Murdoch, senior data-visualisation journalist at the Financial Times, shared how effective data graphics focused on meaning and communication first as opposed to the look of the visual itself. Graphics, he said, could “tell stories” that can impart key research findings in a format that is accessible to the public.

The conference’s official artist, Peter Morey, used scientific drawing and cartoons to share the themes and research findings. Communicating through visual media is “an art of interpretation”, he said. Focusing on key concepts, collaboration with researchers, and a focus on people in the art are all compelling ways to turn abstract numbers and concepts into visual scenes.

Moving forward, the most exciting aspect of this new paradigm is the sense of community and discovery. According to Iain Hyrnaszkiewicz, head of data publishing at Springer Nature, it is “still early days” for discovering the best methods to share and utilise scientific data. There is “still quite a lot of experimentation” about how to increase interactivity and improve the scientific process through better data, leaving room for research groups to explore different ways to share data. While the best ways to move forward are still in development, it is clear data cannot be isolated in emails and lab notebooks. If the goal of good science is to solve problems, meaningful advancements can only emerge from a shakeup in how researchers think about data. The new paradigm is a move to see data as a valuable resource to utilise in solving problems at a scale and speed necessary to address the world’s health, environmental, and social problems.

There is not a single method to accomplish this, but every research group can, and perhaps should, innovate and change how they store, use, and communicate data.

Go to the profile of Jory Fleming

Jory Fleming

Graduate Researcher, University of Oxford

No comments yet.