How leading publishers are embracing open data

The 2022 State of Open Data survey found that the majority of researchers are looking to their publishers for support with sharing their data and making it openly available.

Publishers globally are implementing different strategies and experimenting with new approaches to both support their authors and embrace the tide of open data that is sweeping the research landscape.

With significant policies and mandates in place from the OSTP, NIH, UKRI and other research funders globally, authors are under more pressure than ever to ensure all of their research outputs, not just the final paper, are publicly accessible and shared in a FAIR way.

This, paired with the growing spotlight on robust research integrity and transparency, means that making more of the research process and overall findings as open as possible (or as closed as necessary), is becoming the new normal, or so we hope…

With an increased emphasis on trust and transparency in research and the accelerating concerns around research integrity, in the last couple of years there has been an increase in publishers requiring data availability statements. A 2022 report from Ripeta found that “‘Data available upon request’ as a data availability statement was becoming more prevalent among publications that have been funded by a federal or private agency” and noted that “the use of repositories would greatly improve findability and accessibility of the data.” Therefore, although the need to provide a data availability statement is becoming far more common, the act of responsibly and ‘FAIR-ly’ making that data available is perhaps not as straightforward.

Open data and data sharing is landing highly on the agendas of stakeholders across the research ecosystem. Research institutions, funders and scholarly publishers are all mobilising in different ways as they formulate approaches to the changing landscape and expectations.

It’s also landing highly on the agendas of researchers (and therefore authors) with the burden of publishing research in a FAIR way falling primarily on their shoulders. It’s becoming more and more important for publishers to support their authors as they navigate the complexities of compliance and adopt open practices.

This blog will talk through some of the recent developments we’ve seen in the world of scholarly publishing when it comes to data sharing and showcase some examples of leading publishers that are moving beyond simply requiring a data availability statement and are instead embracing open data, evaluating data sharing practices and experimenting with innovative approaches to encourage data reuse.

PLOS Open Science Indicators 

Understanding the attitudes of researchers towards Open Science and Open Data is extremely important. Figshare partners with Springer Nature and Digital Science annually on the State of Open Data survey and produces a whitepaper report presenting the survey findings. The survey asks researchers to share their opinions on Open Data, their level of awareness of core Open Science principles and information about their own behaviours when it comes to data sharing. In late 2022, Public Library of Science (PLOS), also launched an investigative project of their own that sought to help them better understand the Open Science practices and behaviours of their authors.

As part of their Open Research Solutions programme – the goal of which is to measurably increase adoption of multiple Open Science practices – PLOS announced the launch of a new project: Open Science Indicators.

As a prerequisite to achieving the goal of increasing Open Science practice adoption, PLOS wanted to better understand the needs and practices of the researchers they serve, and so they decided to regularly investigate how Open Science is practised across their content. They extended their partnership with DataSeer AI (who had already worked with PLOS on a similar project for one of their journals, Computational Biology), and decided to investigate levels of three indicators of best practice across all PLOS content from 2019 onwards:

• Rates of data sharing in data repositories
• Rates of code sharing
• Rates of preprint posting, in any preprint server before publication

They worked with DataSeer AI to develop a set of numerical “indicators” linked to specific Open Science characteristics and behaviours observable in published research articles.

The dataset that was produced was made openly available on Figshare, and in their announcement blog, PLOS looked to the future and their plans to add additional dimensions to the data in order to improve and inform strategic planning for the future:

‘At PLOS in particular, we hope that a better understanding of how Open Science tools and practices are applied today can help us to identify barriers, understand community norms,  better support best practices, and track changes over time.’

PLOS is embracing open data as it seeks to more deeply understand the open practices of its author base so that it can offer proficient and relevant support. This project is also of great use to the wider scholarly publishing community, evident in the fact that the first dataset from the project has already been downloaded over 1,000 times and viewed nearly 7,000 times.

We’re very much looking forward to the project progressing. Congratulations to PLOS for spearheading this initiative to better understand the behaviours of authors when it comes to open research practices.

You can view the latest Open Science Indicators Dataset from PLOS here on Figshare and read their blog about the fourth instalment here.

IntechOpen enhancing author services 

As well as investigating author behaviours and planning informed support strategies, it’s encouraging to see leading Open Access publishers such as IntechOpen enhance the level of support they offer their authors when it comes to research data sharing.

In June 2023, IntechOpen Journals announced that they were partnering with Figshare to ‘promote Open Data.’ Motivated by open research principles, IntechOpen had long encouraged its journal authors to make any data and code associated with research papers openly available. Their partnership with Figshare now offers tangible support for their authors to do so.

Enhancing author services was one of the key motivations for implementing Figshare infrastructure into IntechOpen’s submission workflow and integrating it with their submission and review platform ReView, which was built with the intention of seamlessly linking with external data repository platforms such as Figshare.

Along with the motivation to better support authors with the adoption of open principles and provide a seamless submission process, the partnership will also help authors meet requirements of their institutions and funders; ensuring research data is open and FAIR (Findable, Accessible, Interoperable and Reusable).

Initiatives such as this one from IntechOpen Journals highlight a growing commitment from more publishers to work hard for their authors when it comes to data sharing, and offer proficient tools to ensure that Open Data is a very real and achievable goal.

Digital Science Publisher Day: Data Evolution Panel 

On September 6, Digital Science held their annual Publisher Day that brings together our internal experts with external panellists and attendees from the scholarly publishing community.

Session 2 of the day was focused on Data Evolution and saw Figshare’s Founder and CEO give a brief presentation on recent updates to the Figshare platform that are particularly relevant to publishers. This was followed by a fascinating panel discussion on Open Data and data sharing with representatives from Springer Nature, Karger Publishers and PLOS.

Erika Pastrana, Editorial Director, Nature Portfolio at Springer Nature, began the panel session by framing the ‘Data Evolution’ topic in the context of various policies and mandates that have been developing in the space, such as the August 2022 Office of Science and Technology Policy memo, otherwise referred to as the ‘Nelson Memo’ that called for open and equitable public access to all federally funded research.

Christna Chap, Head of Editorial Development at Karger Publishers brought her editorial perspective to the discussion and touched on what the author expectations around support for data sharing from Publishers are and highlighted the importance of consistent and sustained support and outreach from publishers in order to ensure authors are thinking about data sharing long before the time of actual publication.

Lauren Cadwallader, Open Research Manager at PLOS spoke about their mandatory data sharing policy that’s been in place since 2014, and whilst they encourage authors to adhere to ‘best practice’ data sharing, and making it FAIR where possible, they recognise that they don’t want to impose a solution on their authors and instead encourage the overall adoption of Open Science practices that will hopefully be carried forward when publishing with other journals and publishers. They therefore avoid imposing stringent requirements on how and where data is shared when publishing with PLOS, but encourage and promote general Open Science best practices.

The panellists discussed the building excitement and pressures around data being shared in a way that can be useful for machines. There is growing discussion around data reuse and with the growing capabilities of AI tools, it’s perhaps never been more important for data to be as discoverable as possible, to further enhance the possibilities of data reuse writ large. The topic of data reuse led the panellists to a discussion about their shared ambition for the normalisation of data sharing in the future, which could be a key driver to the real ‘opening up’ of science in general.

The discussion was a brilliant summary of what’s landing highly on the agendas of those in the scholarly publishing space working on open data strategies as well as what related trends and expectations they are seeing from their authors.

You can watch the recording of the Data Evolution session here.

‍PLOS Accessible Data button ‍

Policies and guidance around data sharing are increasing and researchers are in many cases required or at least strongly encouraged to make their research data openly available in a FAIR way. For those who are ‘strongly encouraged’, the execution of making data openly available comes down to the motivations to do so.

Greater discoverability of both the full research paper and the research data, citations and receiving credit are all clear motivations for sharing data openly. Droves of views and downloads of Open Data and citations are not guaranteed from the act of sharing data openly alone, although more likely. There are additional considerations to be made when looking to increase the likelihood of reuse, citation and appreciation of Open Data; in particular the clear linking between the data or supplemental material and the research paper itself.

As we’ve described above, there is progress being made in terms of understanding researcher practices, building strategies to support and improve those practices as well as supporting authors with the data sharing itself, but what can the publishing community do to further drive the discoverability of linked data once it’s been shared?

In March 2022, with funding from the Wellcome Trust, PLOS launched an ‘Accessible Data’ initiative that aimed to increase research data sharing and reuse, addressing those popular motivations for data sharing; increased use, citations and credit. PLOS recently published the preliminary results of their ‘Accessible Data experiment.’

The initiative was in itself fairly simple; they added a ‘Accessible Data’ button to published articles that had a linked dataset in Figshare, which highlighted the fact that accompanying data was available and could be explored at the click of a button.

The preliminary results of the first 12 months of this experimental initiative are exciting;

‘Through analysis of 543 Figshare datasets linked to PLOS articles, we observed that the average number of views received per month was 2.5 in the 12 months prior to the launch of the feature and 3.0 in the 12 months following the launch (a statistically significant relative increase of 20%).’

It’s really encouraging to see this creative thinking from leading publishers, in an attempt to drive the discovery and reuse of research data that has been shared. We’re excited to see the progression of this project, as PLOS recently announced that they would be extending the feature to incorporate more journals, more articles and more repositories.

These are just some examples of initiatives from across the publishing sector that highlight a clear engagement with the question of Open Data from leading publishers.

As we continue to accelerate towards mandated data sharing and equitable public access to research, it’s clear that there are multiple ways that publishers can proficiently support Open Data and their authors whilst simultaneously optimising their own content for greater discoverability and reuse, ultimately furthering the progress and potential of research.