Promoting Equitable and Inclusive Implementation of Open Scholarship Policies

Proceedings of a Workshop—in Brief

The Roundtable on Aligning Incentives for Open Scholarship of the National Academies of Sciences, Engineering, and Medicine (the National Academies) brings together stakeholders to discuss the effectiveness of current incentives for adopting open scholarship practices, barriers to adoption, and ways to move forward. On June 26, 2023, the Roundtable convened a virtual workshop to hear from a range of stakeholders on the challenges and opportunities in expanding open scholarship, with a focus on the impact of recent U.S. government initiatives to open the research ecosystem and broaden access to publicly-funded research.

OPENING REMARKS AND AGENDA SETTING

As introduced by Roundtable co-chairs Keith Yamamoto (University of California, San Francisco) and Erin O’Shea (Howard Hughes Medical Institute), federal guidance is introducing major changes in access to and reuse of research and scholarship in the United States. In particular, O’Shea characterized as a “pivotal moment in our community” the August 25, 2022, memorandum Ensuring Free, Immediate, and Equitable Access to Federally Funded Research.¹ The memo (often referred to as the “Nelson memo” because Dr. Alondra Nelson, then acting as director of the Office of Science and Technology, issued it) set a 2025 deadline for federal agencies to make publicly available the research they fund and a 2027 deadline to require the sharing of enhanced metadata, persistent digital identifiers (PIDs), and other information. O’Shea welcomed the opportunity to re-envision and reshape the landscape of scientific inquiry, but she noted the importance of listening to stakeholders affected by the changes. Yamamoto explained the workshop’s goal to discuss ways to proactively align reward systems that support and reinforce open scholarship and to identify actions and actors to reach an equitable and inclusive open scholarship research ecosystem that allows all research enterprise stakeholders to engage in the scientific process. He clarified that the Roundtable considers open scholarship across disciplines, not just the sciences, and that while the Nelson memo is directed to federal agencies, the Roundtable thinks of open scholarship globally.

PERSPECTIVES ON ANTICIPATED IMPACTS OF U.S. FEDERAL POLICy CHANGE

Greg Tananbaum (Roundtable Secretariat; Open Research Funders Group) served as moderator for the workshop, which consisted of eight short presentations from diverse perspectives, including early-career researchers (ECRs), minority-serving institutions, indigenous communities, scholarly publishers, and others. The presentations were followed by discussion with the virtual audience.

__________________

Benefits and Concerns from a Research Administrator

Pramod Khargonekar (University of California, Irvine) spoke as a professor, vice chancellor of research, and former directorate head at the National Science Foundation (NSF). He noted four emphases in the Nelson memo: (1) improved awareness of federally funded research results by all users; (2) reductions in inequities in publishing and access to data; (3) reductions in burdens to comply with federally mandated public access; and (4) strategies to make federally funded publications, data, and other research outputs findable, accessible, interoperable, and reusable (FAIR). Resulting benefits, he suggested, include greater access by non-experts to scholarly research and the potential for innovation, with public use of the NOAA (National Oceanic and Atmospheric Administration) and NASA (National Aeronautics and Space Administration) data as examples.

He also expressed concerns about the memo’s requirements. Costs and administrative burdens will increase, a challenge for institutions that do not have infrastructure or resources to comply with open scholarship policies.² He also pointed to potential conflicts between open access and other constraints, particularly the intersection between open scholarship policies and research security. He urged attention to non-STEM fields. And while compliance with the directive is necessary, it is not sufficient to achieve an inclusive and equitable research ecosystem. “We need to avoid the ‘build it and they will come’ mindset,” he warned. Cultural and organizational barriers must still be addressed beyond policy compliance. He also noted that new technologies such as ChatGPT and developments in machine learning and artificial intelligence would enable us to take advantage of open publication and open data.

“Third-Space” Opportunities from a Student-Run Journal Editor

Ivonne Lujano (Arizona State University [ASU]; Directory of Open Access Journals ambassador for Latin America) noted Latin America’s leadership in open access and science, beginning with the Diamond Model in the late 1990s, can offer useful lessons.³ She also spoke of the tension she has observed between the goals of doctoral education and openness. Students are told to publish in journals with a high impact factor and that are centered in the United States, especially in social sciences and the humanities. The competitive, outcomes-oriented environment in many programs runs counter to open goals, she said. She noted low participation by students as authors, beyond serving as the second or third author in an article, and as reviewers. Most humanities and social science doctoral students think of openness in terms of the Article Processing Charges (APC) model with little awareness of other opportunities.

Lujano shared her experience as editor of the student-run journal Current Issues in Education.⁴ Student-run journals offer a transformational “third space” for graduate students to collaborate with other students and faculty beyond the classroom and research site. They provide students with agency, authorial voice, and collective identity.⁵Current Issues can offer Diamond access because ASU provides support and, except for one paid editor, all are volunteers. The journal is small but has been sustainable for 25 years as an open access publication.

A New Model from a Mission-Driven Publisher

Johns Hopkins University Press has been a leader in open access initiatives, explained Barbara Kline Pope (Hopkins Press). She provided context on current efforts, her perspective of the opportunities and risks of open access, and her opinion about how mission-driven publications can thrive. Hopkins Press’s four divisions encompass journals, a large-scale digital aggregation called Project Muse, books, and a distribution group.⁶ Focusing on journals and Project Muse, Pope said the effort to embrace open access addresses two of the Press’s core values: valuing diversity, equity, inclusion, and access; and taking inspiration from ideas that ignite ingenuity and innovation.

To Pope, open access should become the default mode for scholarly publishing but it must be financially

__________________

responsible. It has become clear that APCs do not serve most humanities scholars as it is hard for them to get funding to pay the charges. In contrast, she offered the subscribe-to-open (S2O) model, which is library-supported and will launch in subscription year 2025.⁷ When a certain financial level is reached, the S2O-participating journals are open to everyone (not just subscribing libraries’ patrons), which she posited “gets us closer to equity.” She suggested an incentive to encourage participation could be a federal mandate, although federal support of humanities research is so limited, it offers little leverage. A more important incentive is competition. Libraries are transitioning from paying for subscriptions to open access, and, in her view, publishers that do not move to sustainable open access will be left behind. Being a first mover represents a strategic advantage to publishers.

She acknowledged the financial risk for publishers whose journal backfiles reside with commercial aggregators as the royalties support their programs. For nonprofit and university publishers to thrive in this changing landscape, Pope said a transition period to mitigate risk is needed because most operate on razor-thin margins. External support is needed. A grant from the Mellon Foundation allowed Project Muse to structure its S2O approach. She concluded with a sense of urgency. “I would hate to think of a world in which all of science is open, but the humanities content remains locked. Open access to all scholarship is one of the best ways to ensure a world in which knowledge enriches the lives of every person,” she said.

Perspectives from the Social Sciences

Simine Vazire (University of Melbourne) explained the role filled by the Society for the Improvement of Psychological Science (SIPS) as a complement to more established societies. She helped start SIPS to develop clearer norms and guidance, including tools, training, and resources to promote transparent and rigorous scholarship. One-third of the SIPS board are ECRs. Annual conferences are oriented toward grassroots action, which has led to many open-access developments in psychology and related fields, including a preprint service for the psychological sciences.

Vazire also offered perspectives on issues around equity and openness as a journal editor. She related that several years ago, a common concern was that moving toward transparency too fast would produce inequity because some researchers would be unable to meet the requirements. However, she observed problems in reviewing articles for publication when some authors share their data and others do not. Once the field is ready and training and resources are available, she advocated requiring transparency while keeping ethical issues, for example related to privacy, in mind. She also called for transparent peer review, in which the content of a review (not the reviewer’s identity) is published, to reduce bias.

She concluded by countering concerns about requiring transparency. First, she clarified that transparency is for information relevant to evaluating research, not people’s lives. She also noted the concern that bad faith actors could misuse the requirement for transparency. She stressed the need for limits on what researchers share to prevent harassment, but not to “close the doors and lock scientifically relevant information away, which will undermine public trust in science more.”

Indigenous Knowledge Sovereignty

Stephanie Carroll (University of Arizona) discussed the interplay between Indigenous Peoples rights and open data. She noted she spoke as a researcher just past the early-career stage, open-access collaborator, and citizen of the Native Village of Kluti-Kaah. “We are concerned about this movement toward open, open, open without ethics and rights perspectives embedded in how we move forward,” she stated. With open science’s goal to make data more equitable, she posed who decides what is equitable, as well as how, when, and where data are made open. Policies need to encompass an ethics and a data infrastructure to ensure that Indigenous Peoples are not forgotten.

Indigenous data are “data, information, and knowledges, in any format, that impacts Indigenous Peoples, nations and communities at the collective and individual levels.”⁸ Sovereignty of those data must be embedded into open access, she said, pointing to the Indigenous Data Sovereignty Movement that began in 2016 and has been

__________________

incorporated into global policies and the policies of many nations. While there is no U.S. national policy, there are many Tribal policies around research and data-sharing.

In 2019, the Global Indigenous Data Alliance (GIDA) was launched to “put Indigenous data back into Indigenous hands” through publications, policy development, and other efforts to create norms and standards. Indigenous scholars and allies looked at existing open frameworks and principles, most notably FAIR,⁹ and developed the CARE principles to work in tandem with them. CARE principles call for (1) Collective benefit; (2) Authority to control; (3) Responsibility; and (4) Ethics.¹⁰ She noted that CARE is not a uniform standard but sets a minimum for those interacting with Indigenous data. GIDA has also set out 12 rights as expressed by Indigenous Peoples about data and governance.¹¹ Carroll shared several examples of implementing these principles and rights, including acknowledgment of Indigenous Data Sovereignty in the Data Management and Sharing Policy of the National Institutes of Health (NIH) and a movement to incorporate recognition of Indigenous Peoples rights within the enrichment of metadata. Changes are needed not just in ethics but infrastructure, she pointed out, and around collective, not just individual, rights.

Equity and Inclusiveness through the Lens of Historically Black Colleges and Universities

Reflecting on open scholarship and its impact on equity and inclusiveness from an HBCU lens, Vicki Coleman (North Carolina A&T) underscored her support of the Nelson Memo and of 2023 as the Year of Open Science. But, as the dean of libraries at an R2 institution with limited resources, “reality soon started to set in.” She noted the work needed to act on the changes, for example to comply with the NIH data policy mentioned by Carroll. With limited infrastructure, staff is pulled in many different directions. She is now trying to figure out how to handle the new requirements with a total library staff of 40 full-time-equivalent employees. She welcomed the

inclusion of allowable costs for data management in the policy but noted it does not cover hiring staff.

Moving to publications, she discussed the opportunities and barriers of open access for R2 institutions. Despite publishers’ claims of transparency related to APCs and other charges, she pointed out ambiguity, such as learning that one journal charges fees ranging from $200 to $10,000. She cannot cover APCs in her library budget, but she works with authors to ensure a journal is not predatory and seeks discounts through consortia and other means. However, she stressed, APCs can be a barrier to open access.

Policies can be open but not inclusive and equitable, especially when contrasting repositories in the Global North with the Global South, as well as within the United States. Coleman called attention to the U.S. Repository Network, an inclusive network committed to advancing all open repositories.¹² She urged attention to the impact of inequity and years of underfunding of HBCUs. When agencies roll out policies with uniform requirements under the assumption that all institutions have equal access to resources to implement them, HBCUs and others with lower resources face barriers to comply, Coleman concluded.

Impacts of Openness on Early-Career Researchers

Gwen Chodur (University of California Davis; UC Graduate Professional Council) noted her involvement in efforts to advocate for high-quality mentorship and more equitable access to materials and education across R1 and R2 institutions. She discussed the impact of policies on ECRs.

Although issues in the research enterprise, such as the reproducibility crisis, undervaluing of graduate students and post-docs, and growing concern about APCs and predatory journals, will not be fully solved by open access, she said, open access can address them. Open access allows ECRs to take risks and mitigates publication bias, she suggested. They can ask more ambitious and interesting questions, move science forward, demonstrate technical expertise, and make their work more visible to a larger audience. Publication delays have a disproportionate impact on ECRs, but being able to point

__________________

to work in a repository can help in the competitive search for funding. Another benefit is the development of pilot data.

However, Chodur highlighted that ECRs are vulnerable in some open-access scenarios, especially those at smaller institutions. It becomes easier for better funded or more established labs to take advantage of shared data and move it forward. She urged standards for open access repositories, such as for citations and authorship. It is important to develop and enforce guidelines for critical review so as not to reinforce power differentials and “punch down ECRs and junior faculty.” She suggested better training in reproducible methods in all graduate programs and training.¹³ Mentorship standards should be developed, noting some cases of retracted articles in which senior scientists blame graduate students for data problems. She also urged changing the culture around science to incentivize open access publications and open data, even when not required. Consensus standards for how to credit work that has been placed in a repository but has not gone through peer review should be established, such as through citation, incentives to collaborate, or supplemental funding awards. Finally, Chudor called attention to the rights around publicly open data, both for participants and those who generate it. The Prospero registries could serve as a valuable model, she suggested.¹⁴

Open Science across a University Campus

As the final presenter, Russ Poldrack (Stanford University) spoke from the perspective of a cognitive neuroscientist working to improve the reproducibility of neuroscience and as the leader of an effort to extend open scholarship across campus. Openness is key to reproducibility, he reiterated, and a researcher-led initiative to share brain imaging, along with other efforts, has been valuable in his field. He noted academic researchers can take more risks than those within federal agencies.

The Center for Open Reproducible Science (CORES), which has support from a different part of the campus, serves as Stanford’s liaison to the Higher Education Leadership Initiative for Open Scholarship (HELIOS) and other efforts to align incentives around open scholarship. Federal actions such as establishing 2023 as the Year of Open Science demonstrated the timeliness and importance of these efforts to university leadership. The new requirements motivate researchers to learn how to engage in open scholarship to move towards open science. The guidance has also been useful as leverage to push against overly restrictive data-sharing or IP policies at many institutions.

CORES developed a “gentle on-ramp” guide to researchers called Open by Design.¹⁵ They offered their materials to NASA’s Transform to Open Science and hosted a NASA “Open Science 101” workshop. Acknowledging that Stanford has more resources than many institutions, he stressed that the effort has taken hard, grassroots work to be successful. Engaging with trainees has built community. He observed that ECRs across institutions see open science as key to building their careers, although the challenge is when the incentives for hiring and promotions are not aligned with their passion, and traditional metrics of productivity and impact factors prevail.¹⁶

DISCUSSION

Closing the “Last Mile”

To launch the discussion, Tananbaum asked about attaining the last mile: that is, to connect open scholarship to public engagement with, and confidence in, science. Pope agreed that just making research available is not enough. The evidence must be accessible for those who want it, but tools are needed to make the content relatable to audiences, which she acknowledged is easier said than done. She suggested that AI may help. To connect with the public, Carroll said she promotes using science in the co-production of policies and practices in universities, agencies, and other institutions and communities. Lugano called for reforming the reward system to recognize outcomes that are not just focused on the academic community, such as podcasts or arts-related efforts. Chodur suggested that open science could be a precursor to an increased focus on science education and literacy.

__________________

Training

Several attendees asked about training people at different stages in their careers, including open and equitable access. Tananbaum noted a HELIOS working group is working with the plethora of existing resources to make them easier to navigate. Poldrack acknowledged the technology can be intimidating. User research with people who are coming to open science with interest but little knowledge can inform efforts to tailor relevant training. Tananbaum noted the value of professional societies to provide discipline-specific norms and practices, led by the 11 heterogeneous societies and associations of the Alliance for Open Scholarship (All4OS). Participants from two societies explained how they provide guidance and support for their members to meet federal guidelines, factor in ethics and equity, and deal with the realities of differential resources.¹⁷

Transition to Alternative Dissemination Models

In an era with tight library budgets and small margins for noncommercial publishers, the willingness to experiment with new models is not optimal, Tananbaum noted. Vazire commented that the society that publishes the journal she edits is considering moving from APCs toward diamond access—free to both readers and authors. She said she sees the benefits of this model, but also understands the reliance on APC-generated funds to pay for staff and other expenses.

Several librarians weighed in. A participant shared that MIT ended its contract with Elsevier in 2020 and instead pays a fee per article. A few faculty say it is marginally less convenient, but the costs are one-tenth the previous contract. The savings will be invested into other open scholarship efforts, she said. A participant from the University of Louisville noted the significance of impact factor in research and suggested the need for a high-level change agent in transitioning to a new model, which requires analysis of how to get people on board and allocate resources. Coleman said tough decisions are required by library deans and directors, especially when it seems there is little support for change. She added that library

staff vacancies provide an opportunity to re-think job responsibilities and budgets to support open scholarship.

From a publisher’s standpoint, the willingness to experiment with non-APC models depends on the market being receptive, Tananbaum commented. Pope said the S2O model she described is one possibility, and others will probably be developed as more is learned. Publishers differ in their perceptions of risk and ability to take those risks. Sharing the risk and trying to get the risk funded during a transition period, will help grow the number of publishers involved in the transformation to open access, she asserted.

Balancing Indigenous Data Sovereignty and Open Data

Carroll said it is important to recognize the tension between Indigenous Data Sovereignty and open access.¹⁸ Previous practices collected a lot of data about Indigenous Peoples and individuals without making much of those data accessible to them. She noted the CARE principles have been well received, and that training on the use of Indigenous data is critical. She also stressed the Indigenous Data Sovereignty Movement is one of many movements that advance rights to achieve synergy. Tananbaum noted the tie-in with public confidence in science.

Implementation of the Nelson Memo and Agency Policies

In considering how to make compliance with the new policies easier for institutions, Poldrack said that training for understanding the new policies and sharing of ideas will help, but meeting the requirements will take time and effort, especially at the beginning. Carroll stressed that training be designed and delivered in a way that recognizes cultural, ethical, legal, and social input, perhaps through a badging system for trainers or through The Carpentries.¹⁹ Chodur welcomed the availability of administrative supplements but expressed concern that the additional burden will fall on senior graduate students and post-docs.

Tananbaum referred to a comment made by Roundtable co-chair Yamamoto at a previous Roundtable: Open scholarship is not a distraction from the science, it is

__________________

the science. It must be integrated into what it means to be a scientist or researcher. Thinking of incentives to make this a reality, Poldrack suggested the positive effect if people get credit for their efforts. Vazire suggested awards given to projects, not just individuals, to encourage openness, as well as recognition of tool development and research outputs, not just articles, on professional CVs. Tananbaum noted the Open Science Toolkit provides a rubric to evaluate an individual’s contribution to open scholarship that can be used as in input in hiring, promotion and tenure.²⁰

The Connection with Better Scholarship

A participant commented about the assumption that open scholarship leads to more inclusive and rigorous processes and outcomes and inquired how to know if this assumption is correct. Chodur commented on the average 10-plus year lag in biomedical science between when something appears in the literature and when it is practiced in the clinic. She posited that open science can perhaps narrow that lag. Vazire brought up emerging research from the field of metascience to study this question.

CONCLUDING REMARKS

O’Shea thanked participants for their range of insights. Yamamoto reflected on some themes he heard: Proper alignment with the Nelson memo will require far more than making papers free to read. Activities and behaviors must be rewarded to mutually reinforce vectors of change. Operational work is needed related to such issues as reuse rights, machine readability, infrastructure, dissemination workflows, and more to move from a policy pronouncement to implementation. Picking up on this point, O’Shea stressed the role of the Roundtable to grow a coalition committed to working on these issues and to identify opportunities to lean into collective action to de-risk the transition to open by default.

__________________

DISCLAIMER This Proceedings of a Workshop—in Brief was prepared by Paula Whitacre as a factual summary of what occurred at the workshop. The planning committee’s role was limited to planning the workshop. The statements made are those of the rapporteur or individual workshop participants and do not necessarily represent the views of all workshop participants; the planning committee; or the National Academies of Sciences, Engineering, and Medicine.

REVIEWERS To ensure that it meets institutional standards for quality and objectivity, this Proceedings of a Workshop—in Brief was reviewed in draft form by Alicia Salaz, University of Oregon and Erik Schultes, GO FAIR Foundation. The review comments and draft manuscript remain confidential to protect the integrity of the process.

PLANNING COMMITTEE Heather Joseph (Chair), Scholarly Publishing and Academic Resources Coalition; Chris Bourg, Massachusetts Institute of Technology; and Maryrose Franko, Health Research Alliance.

STAFF Thomas Arrison, Director, Board on Research Data and Information (BRDI); Greg Tananbaum, Roundtable Secretariat (until December 2023); George Strawn, Scholar, BRDI; and Emi Kameyama, Program Officer, BRDI.

SPONSORS This workshop was supported by the John Templeton Foundation, Robert Wood Johnson Foundation, and Open Research Funders Group. Over its first three years, the roundtable has also received support from the Arcadia Fund, Arnold Ventures, Schmidt Futures, the Leona M. and Harry B. Helmsley Charitable Trust, the National Library of Medicine, the Open Research Funders Group, the Open Society Foundations, and the Wellcome Trust.

For additional information regarding the workshop, visit: www.nas.edu/brdi.

SUGGESTED CITATION National Academies of Sciences, Engineering, and Medicine. 2024. Promoting Equitable and Inclusive Implementation of Open Scholarship Policies: Proceedings of a Workshop—in Brief. Washington, DC: The National Academies Press. https://doi.org/10.17226/27801.