5 Steps To Encourage Data Sharing In Scientific Research

Data sharing among scientific researchers is a crucial aspect of innovation as it helps to spread awareness not only of successes but also failures.  Encouraging the practice of data sharing is not always easy however, not least because the citation process tends to encourage researchers to think about themselves first of all.

This is bad, as data sharing has been proven to make researchers considerably more productive, producing higher quality work that is cited more frequently, thus providing a benefit not only to themselves, but to society more generally.  For instance, something as simple as data archiving can double the publication output of each research project, whilst boosting citations by up to 50%.  The European Commission have even put the cost of poor data sharing at around €10 billion per year.

A new paper from Springer Nature explores how researchers can be encouraged to share data, and comes up with five key factors they believe are required.

“While we continue to see researchers increasingly sharing data, the majority of the research community are not yet managing or sharing data in ways that make it findable, accessible or reusable. The utopia of findable, accessible, interoperable and reusable (FAIR) data is still some way off,” the authors explain.

Encouraging data sharing

The paper outlines five factors that they believe are key to the acceleration of data sharing in the research community:

  1. Clear policy, whether from funders, institutions, journals or any of the other stakeholders in the research community.  A clear requirement for data management will be a great first step to encourage data sharing.
  2. Better credit, so that data sharing is something that clearly rewards researchers, whether through formal recognition, career advancement or inclusion in research assessments.
  3. Explicit funding for data management and data sharing tasks.  Policy is a vital first step, but it needs to be backed by money to be truly valuable.
  4. Practical help for researchers in areas such as data management, data searching and even the act of sharing data itself.  It’s naive to assume that researchers have data science skills.
  5. Training and education can then fill in some of these gaps to help researchers build clear data science skills and capabilities.  Communicating the benefits of data management is also an important part of this process.

“None of these essential factors can be solved by one stakeholder alone: we must act together, and we must act now, to encourage data sharing across discipline and geographic boundaries. Support from all stakeholders – funders, institutions, publishers, and the wider research community – could make all the difference,” the authors conclude.

Facebooktwitterredditpinterestlinkedinmail