Facilitating the reproducibility and replicability of published research is a growing issue in scholarly publishing with a concurrent trend in open data mandates from funders, institutions, and publishers. Recognizing that the availability of research data plays a vital role in ensuring reproducibility, the International Association of Scientific, Technical and Medical Publishers (STM) has declared 2020 as STM Research Data Year. The dedicated action plan for the initiative follows the theme Share – Link – Cite and strives to (1) increase the number of journals with data policies and articles with data availability statements, (2) expand the number of journals depositing data links to the SCHOLIX framework, and (3) grow the volume of citations to datasets while adhering to the Joint Declaration of Data Citation Principles, created by the Data Citation Synthesis Group.
“There is a growing need to ensure the findability, accessibility and re-usability of research data for all stakeholders working across scholarly communications,” STM’s CEO, Ian Moss, said in a statement. “Successful efforts by some publishers have already shown that boosting the sharing of data helps to ensure the long-term integrity of research, whilst also continues widening publisher support for Open Science.”
The Research Data Year action plan focuses on sharing best practices and creating materials to help publishers accelerate research data implementations. A getting started webpage and an additional resources webpage have been launched to aid publishers. STM plans to organize workshops and webinars where publishers can exchange experiences, tips, and lessons learned
and will also work with individual publishers, including onsite visits that support implementation efforts.
Research Data Year is a collaboration of 11 publishers that, according to Joris van Rossum, STM’s research data director, publish more than half of all journal articles. However, Einar Ryvarden, the digital manager of The Journal of The Norwegian Medical Association, an independent scientific journal with a staff of 20, argues that without using a platform partner or being part of a large publishing company, building and using data exchange services are costly processes. While well-established journals and societies, such as the 11 Research Data Year Publisher Collaborators, have internal resources to understand, maintain, and experiment with new demands such as open data mandates, smaller and younger publications may struggle because of the lack of financial resources and the staff’s inability to maintain industry-standard technology and processes. Curating, metadata preparing, checking, reviewing, and publishing datasets and then ensuring that the repository links work in the article can take weeks of dedicated effort─weeks that smaller, independent journals do not have or cannot spare, Ryvarden maintains.
He suggests that instead of journals taking on this additional workload, data should be fetched by third parties. “Economically, it’s a simple choice,” Ryvarden said. “Let a handful of indexing and data collectors change their input systems instead of having a whole industry try to agree and then change thousands of systems.” Similar to how Google and other search engines work, sites could be crawled to retrieve the data needed, he explained.
As an open access journal that understands and supports the importance and trend toward open science, the Ochsner Journal staff will continue to research and consider practical implementations for promoting reproducibility and replicability of published research.