Monday July 26th 2:00pm ET – Office Hours for PLOS/CRL/NERL unlimited publishing offer: Final Session Please join PLOS’ Partnerships team for an…
Building Trust to Break Down Barriers
By Iain Hrynaszkiewicz, Publisher, Open Research, PLOS
Note: the following perspective was published as part of Digital Science’s annual survey and report, The State of Open Data 2019 , to coincide with global celebrations around Open Access Week.
The biggest barrier to research data sharing and reuse seems to be a matter of trust, and in particular trust in what others may do with researchers’ data if it is made openly available. The 2019 State of Open Data survey revealed that more than 2,000 respondents had concerns about misuse of their research data.
Concerns about data misuse represent a multitude of issues; fears that errors could be found in their work, or that the data could be misinterpreted or research participant privacy be compromised. Researchers might also be concerned that their data will be reused for purposes they did not intend, such as commercial exploitation, or for misleading or inappropriate secondary analyses.1
The 2019 survey provides insights from one of the largest pools of respondents ever, but this particular barrier – concerns about misuse of data – should not come as a surprise. It is supported by several previous surveys which, when combined, total many thousands of researcher opinions.1,2,3,4 With such compelling evidence that these researcher concerns exist, there is an ever-more important need to explore why they persist.
Practical problems, including where, how and when to share research data, have received much attention.5,6 However, concerns about misuse, alongside the potential to lose publication opportunities (“fear of being scooped”), or going against the more conservative data sharing culture in their own field of research often feature amongst researchers’ top concerns.
Features of technological solutions, such as the ability to share data privately in repositories before publication, and policy-level support for researchers’ reasonable first use of their data8 go some way to addressing this issue of trust. However, trust is more a matter of culture than technology. With repositories being used by around a quarter of researchers9,10 investing in people rather than infrastructure may be a more pressing issue to change research culture, as Dr Marta Teperek, who coordinates one of the largest institutional data stewardship programmes at TU Delft in the Netherlands, has concluded.11
Investing in skills and training for individual researchers is one possible solution, although Professor Barend Mons has argued that data stewardship, like computer programming, is too specialist a task to expect all researchers to undertake. He recommends that universities provide one data steward for every 20 researchers.12 Like most difficult problems, there is rarely a single solution. Experimentation and collaboration are essential if we are to enable researchers to build greater trust in the power of data reuse to advance science – in the same way that the unbridled reuse of content helped grow the world wide web so rapidly.
Scholarly publishers have made good progress in the first step towards changing research culture to support data sharing by raising awareness through the implementation of journal research data policies in the last five years.
At PLOS we have invested significantly in people and processes to support a strong journal data sharing policy since 2014.13 We are seeing a steady increase year-on-year in the proportion of PLOS authors who use a data repository.14 Although less costly for publishers, journal policies that only encourage data sharing have much lower levels of compliance.15
With so many papers subject to a journal data sharing policy now published it is possible to explore actual data sharing behaviours, and the benefits of sharing, at a large scale. A study of more than half a million PLOS and BMC papers found that researchers who stored their data in a repository were associated with an average 25% increase in citations to their research papers.16 This is important given that the 2019 survey suggests that a full citation to research papers remains the strongest incentive for researchers to share their data.
While more publishers must invest further in data sharing support for researchers, we should be encouraged by collaborations between publishers and other stakeholders to enable data sharing to be more effective and rewarding. In 2019 the STM Association is collaborating with a Research Data Alliance initiative to implement consistent journal data sharing policies.17 There are indications from some large publishers that they will focus on strengthening journal policies, signalling the importance of a greater commitment than only encouraging data sharing, or “data available on request”. As publishers, we can help to build trust with researchers by being open ourselves: open with our content; our data; our policies; and open with our own data sharing insights.
There are undoubtedly costs associated with implementation of strong policies and solutions for data sharing but, to paraphrase Dr Jean- Claude Burgelman at September 2019’s CODATA Beijing conference, open science is ultimately an investment, not a cost.18 Open science, and indeed open research, is an investment in creating more reliable and reusable knowledge for the future. While data sharing culture and fears over misuse of data persist in today’s research environment, with so many organizations invested in finding innovative solutions to the problems that prevent optimal data sharing, we can remain optimistic that these barriers will ultimately be broken down in the future.
1 Wiley Open Science Researcher Survey 2016 [Internet]. [cited 15 Nov 2018]. Available: https://figshare.com/articles/ Wiley_Open_Science_Researcher_ Survey_2016/4748332/2
2 Science D, Hahnel M, Fane B, Treadway J, Baynes G, Wilkinson R, et al. The State of Open Data Report 2018. 2018
3 Allagnat L, Allin K, Baynes G, Hrynaszkiewicz I, Lucraft M. Challenges and Opportunities for Data Sharing in Japan. Figshare. 2019; doi:10.6084/ m9.figshare.7999451.v1
4 Lucraft M, Allin K, Baynes G, Sakellaropoulou R. Challenges and Opportunities for Data Sharing in China. Figshare. 2019; doi:10.6084/ m9.figshare.7326605.v1
5 Lucraft M, Baynes G, Allin K, Hrynaszkiewicz I, Khodiyar V. Five Essential Factors for Data Sharing. Figshare. 2019; doi:10.6084/ m9.figshare.7807949.v2
6 Stuart D, Baynes G, Hrynaszkiewicz I, Allin K, Penny D, Lucraft M, et al. Whitepaper: Practical challenges for researchers in data sharing [Internet]. 2018. Available: https://figshare. com/articles/Whitepaper_Practical_ challenges_for_researchers_in_data_ sharing/5975011
7 Houtkoop BL, Chambers C, Macleod M, Bishop DVM, Nichols TE, Wagenmakers E-J. Data sharing in psychology: A survey on barriers and preconditions. Advances in Methods and Practices in Psychological Science. 2018;1: 251524591775188. doi:10.1177/2515245917751886
8 Concordat on Open Research Data [Internet]. 2016. Available: http://www. rcuk.ac.uk/documents/documents/ concordatonopenresearchdata-pdf/
9 Science D, Hahnel M, Treadway J, Fane B, Kiley R, Peters D, et al. The State of Open Data Report 2017. 2017;
10 Open Data: the researcher perspective – survey and case studies [Internet]. 4 Apr 2017 [cited 15 Nov 2018]. Available: https://data.mendeley.com/ datasets/bwrnfb4bvh/1
11 The main obstacles to better research data management and sharing are cultural. But change is in our hands | Impact of Social Sciences [Internet]. [cited 24 Sep 2019]. Available: https://blogs.lse.ac.uk/ impactofsocialsciences/2018/11/14/ the-main-obstacles-to-better-research-data-management-and-sharing-are-cultural-but-change-is-in-our-hands/
12 Popkin G. Data sharing and how it can benefit your scientific career. Nature. 2019;569: 445–447. doi:10.1038/ d41586-019-01506-x
13 PLOS’ New Data Policy: Public Access to Data | EveryONE: The PLOS ONE blog [Internet]. [cited 28 Mar 2019]. Available: https://blogs.plos.org/ everyone/2014/02/24/plos-new-data-policy-public-access-data-2/
15 Federer LM, Belter CW, Joubert DJ, Livinski A, Lu Y-L, Snyders LN, et al. Data sharing in PLOS ONE: An analysis of Data Availability Statements. PLoS ONE. 2018;13: e0194768. doi:10.1371/journal.pone.0194768
16 Vines TH, Andrew RL, Bock DG, Franklin MT, Gilbert KJ, Kane NC, et al. Mandated data archiving greatly improves access to research data. FASEB journal : official publication of the Federation of American Societies for Experimental Biology. 2013; fj.12- 218164-. Available: http://www.fasebj. org/content/early/2013/01/07/fj.12- 218164
17 Colavizza G, Hrynaszkiewicz I, Staden I, Whitaker K, McGillivray B. The citation advantage of linking publications to research data. arXiv. 2019; https://arxiv. org/abs/1907.02565
18 Cost-benefit analysis for FAIR research data – Publications Office of the EU [Internet]. [cited 24 Sep 2019]. Available: https://publications.europa. eu/en/publication-detail/-/publication/ d375368c-1a0a-11e9-8d04- 01aa75ed71a1