Open Data in Global Environmental Research: The Belmont Forum’s Open Data Survey

The findings of the Belmont Forum’s survey on Open Data – published on PLOS ONE - demonstrate a substantial uptake of data sharing through e-infrastructures and a further need for enhancement and consolidation. 98% of respondents claimed publishing of open data to be important in their community to advance research!


The Belmont Forum is a group of high-level representatives from major funding agencies across the globe. It coordinates funding for collaborative research actions to address the challenges and opportunities of global environmental change.

To collaboratively develop capable e-infrastructures that can meet the data arising from the Belmont Challenge and the Future Earth agenda, a multi-phased E-Infrastructure and Data Management Collaborative Research Action (CRA) was initiated in late 2013. The main goal of this initiative is to develop a strategy and implementation plan to further shape strategic science policies, outlining what can be done better, in a multilateral way, to fund and support global environmental change research.

In pursuit of this goal, the Belmont Forum working group on open data (one of 6 working groups under this CRA) developed a survey inviting researchers of various science communities and others who are providing and/or using open data in the scope of global environmental change, to share their views and experiences on data publishing, access and (re)use. Nineteen questions of the survey were distributed through targeted (disciplinary and professional) mailing lists.

Findings: perceived properties and attributes of open data

From 16 September to 12 November 2014 Belmont Forum’s Open Data Survey collected 1330 (used those 1253) responses (from 80 countries). The majority of responses came from central Europe and the United States.

The answers to the question “Which attributes do you think are most important to open data?” highlight the importance of “The ability to assess the quality, to select based on metadata, and to easily access and (re)use the data” (n = 944 to 973 responses).

3/5 of all respondents considered  idea of openness as desirable attributes, while 2/5 of them still considered the option to restrict access (“mandatory openness is not an option”) as a very important attribute keeping control of the release date of data.

A “Public Domain” or “Attribution” license were considered most useful for open data (n = 712 to 820 responses).

175 respondents provided a wide range of examples of known data publishing guidelines, including licenses, data policies, repository/archive and e-infrastructures, declarations, handbook, general information, government data, data publishing and data journals, technical guides, literature.

“Four out of five respondents highlighted that open data is crucial for advancing research (n = 853 to 878 responses)”.

“The commitment to publish data as open data seems to be driven by research-intrinsic motives, combining general and personal motivations (n = 834 to 861 responses)”.  Other main motivators to publish data as open data are: funder-endorsed data policies, scientific merits and accelerating research and applications. These issues of motivation need to be identified and addressed.

Discovery of open data versus enablers and expectations for sharing data

“References in journal articles, web search engines and data repositories were identified as the most common discovery routes (n = 774 respondents selected at least one option)”.

Trusted data repositories could further ease this process by providing technical and administrative support for researchers”.

“Paying for data as well as varying data quality, standards and formats were considered least acceptable when accessing data (n = 687 to 731 responses)”.

“Core expectations of users of data infrastructure were that attribution information is provided and that data is citable (n = 890 to 911 responses)”.

“The release of data was seen as a secondary step compared to publishing results (n = 825 to 854 responses) […]

The desire to publish results before releasing data was somewhat more prevalent at early stages of a research career [Moreover] The desire to keep control of their data seemed to be more prevalent for some disciplines”.

Many authors have promoted the benefits of data sharing which might have impacted the research community which responded to our survey”, reported the authors of the article.

Actions recommended to the Belmont Forum

Funders should make open data archiving mandatory”, while ethics of data sharing and re-use should be taken into account when proposing guidelines for open data sharing and re-use”.

“Support and training activities should be supported in concerted ways”.

Interoperability between infrastructures should be further facilitated”.


Source: Schmidt B, Gemeinholzer B, Treloar A (2016) Open Data in Global Environmental Research: The Belmont Forum’s Open Data Survey. PLoS ONE 11(1)

 

Are Belmont Forum’s outcomes enough to strengthen the case for data sharing globally?

Do you think something is still undervalued?

We would appreciate hearing your opinion!