Sharing research data: trends in journal publishing

Online journals do not have the same space restrictions as print, and are, therefore,  able to include supplementary materials (e.g. tables, power-point, images, videos, and other files). At the same time, there is a growing movement for open access to research data, with research funders in the UK stipulating policies on research data.

The open access biomedical  journal, F1000 Research, has, since its inception, had a policy in place to require authors publishing with them to make available the research data relevant to the article. They strongly believe it is important for others to have access to the raw data so that it is possible to verify and replicate the results.

In recent months, we have seen:

Dec 2013 – Mar 2014: PLOS , announced a revised data access policy in a Dec 2013 blog post. This was really a clarification of the policy that they have always had  to encourage access to the raw data used in research articles. The new policy stated that  from March 2014, authors publishing in PLOS journals would have to provide a “data availability statement” and that authors must be willing to share research data. Allowances are made for sensitive data and ethical standards must be applied.  This set off a debate amongst researchers  for and against data sharing, and the PLOS policy became confused. (See the Mar 2014 Blog post which aims to clarify the position.) Like F1000 Research, PLOS believe that access to the raw data used in the research must be available in order for others to test the reliability of results and to build upon initial research.

Jan 2014: Taylor &  Francis (publishers of some 1800 journals) announce that henceforth all articles will have supplementary data uploaded to figshare. This can include media files and research data. All articles will show a link to the related data on figshare, and there will also be a widget to allow readers to view the data alongside the article text.

Mar 2014: Nature announce the launch in May 2014 of an open access journal, Scientific Data. This is, they say, “a new open-access, online-only publication for descriptions of scientifically valuable datasets” and is designed to help researchers deal with the frequent demands of funders to provide data management plans and open access to research data.  This initiative is supported by DRYAD, FIGSHARE, BIOSHARING and ISATOOLS.

http://www.lumaxart.com

Some of the arguments:

1)      Proper assessment of published research can only take place where there is access to the raw data upon which the results depend.

2)      This isn’t feasible or ethical in all subject fields, e.g. where there are issues of confidentiality.

3)      Accessibility of research data sets will help speed  the progress of scientific research.

4)      Proper data management and accessibility will give credit to the originators when it is reused by other researchers, and may even lead to increased citations.

5)      Funding bodies want  research data to be made available (where feasible) for other researchers to mine and re-use, and publisher policies like this may be helpful to researchers in complying with these conditions.

6)      Access and re-use policies will give better return on research funders’ investments.

7)      When the dataset is linked to the article, and thus to the authors, this will increase the likelihood of correct attribution to the originators of the data.

Further reading on data sharing…

Borgman, CL. The conundrum of sharing research data. Journal of the American Society for Information Science and Technology. 2012;63(6):1059-78. 

Peer, L. Participant confidentiality and open access to research data. LSE Impact of Social Sciences Blog. 12 Feb 2013

Piwowar H, Day R, Fridsma D. Sharing detailed research data is associated with increased citation rate. PLoS One. 2007;2(3).

Wicherts JM, Bakker M, Molenaar D. Willingness to share research data is related to the strength of the evidence and the quality of reporting of statistical results. PLoS One. 2011;6(11).