3
Roundtable Priorities for Advancing Open Science
Workshop participants broke into several small group sessions with the goal of suggesting possible strategies for the Roundtable on Aligning Incentives for Open Science to catalyze changes to research incentivization systems at scale. Greg Tananbaum of the Open Research Funders Group moderated a session in which a representative of each of the breakout groups summarized the group’s discussion. Five priorities emerged from those summaries for the roundtable to consider that could help bring incentive structures into alignment with open science practices:
- Ensure researchers have the guidance, training, and resources to fully participate in open science practices.
- Encourage deans, presidents, and provosts to signal that open science is a priority.
- Make the reporting of open activities and the accrual of credit both easy and normative.
- Build upon policies and processes already in place to support open science.
- Identify and address the financial and human costs of open science.
(1) Ensure researchers have the guidance, training, and resources to fully participate in open science practices. Susan Fitzpatrick of the
James S. McDonnell Foundation said that transitioning to a new system of research dissemination carries a degree of uncertainty and risk for all parties. This can be mitigated through proactive measures such as aligning the effort with clear incentives for researchers, leveraging existing institutional infrastructure to socialize and support new norms, and coordinating with funders and professional societies to develop discipline-specific guidance.
The roundtable should work to systematically ensure scientists, especially those working at universities, are aware of the available resources at their disposal, on campus and beyond, said Derrick Anderson of Arizona State University. Researchers can learn from exemplars in their disciplines who are already successfully practicing and championing in open science activities. Michael Dougherty of the University of Maryland discussed the importance of devising new training and professional development programs for students, faculty, and staff on open science tools, methodologies, and approaches as gaps in these available resources and case studies are identified. Fitzpatrick also noted that funding will be necessary support to foster open tools and practices. Giving visibility to institutions and departments providing training in open scholarship and demonstrating the impact could be beneficial as well, said Jennifer Gibson of eLife. Russell Poldrack of Stanford University added that there is a need to develop tools to mentor people at the beginning of their careers to adopt open science practices and give professional recognition for sharing data and preprints.
Participants also discussed assessing and reevaluating the role of open science as part of the faculty workload, including creating incentives and rewards. Robert Hanisch of the National Institute of Standards and Technology noted that faculty are expected to drive change in open science and yet are not recognized for their efforts through the current promotion and tenure process. To address this, Dougherty suggested consistent language in faculty job ads and annual review criteria. This would ensure that researchers are encouraged to pursue open science activities and practices as a core element of their workload. Gibson agreed that there is a need to recognize open science and collaboration or team science in tenure and promotion processes, which could perhaps be accomplished in part by highlighting open data citations on CVs. Maryrose Franko of the Health Research Alliance said providing guidance and examples of wording for applications could help job candidates convey the impact of open science in their work. This language could be highlighted in promotion and tenure packages and grant applications.
(2) Encourage deans, presidents, and provosts to signal that open science is a priority. Breakout session participants emphasized the critical role that senior leadership can play in signaling the importance of open science at their institutions. A concerted effort to bring campus and institutional leaders on board with open science is key if open science is going to be viewed as valuable and rewarded, said Dougherty. One way to promote open science could be to host workshops with academic consortia, such as the Big Ten Academic Alliance.
Boyana Konforti of the Howard Hughes Medical Institute reiterated the importance and role of university leaders embracing open science, noting that broader communication with faculty and administrators at the college level is needed to link the movement with policies. Katie Steen of the Association of American Universities also discussed how university leaders can use high-level signaling statements to communicate the value of open research broadly across campus. Chris Bourg of the Massachusetts Institute of Technology added that leadership can build upon those signals by supporting the infrastructure necessary for open science and scholarship.
University leaders can also play a role responding to structural racism, including connecting this movement to open science policies, said Konforti. Historically Black Colleges and Universities can play a strong role. To this point, Bourg added that leaders can promote a broader understanding of the role open science plays in enabling research excellence and inclusivity.
(3) Make the reporting of open activities and the accrual of credit both easy and normative. Some of the breakout groups explored the role the roundtable could play in promoting an easy, standard way for researchers to report open activities and incorporate open science into their workflow. Ashley Farley of the Bill & Melinda Gates Foundation noted that there is a need to measure the impact of open science and collaboration so that these activities might be properly rewarded. Agreement on open mandates, including dedicated funding, could accelerate this effort. Employers could also reward those who participate in open practices, demonstrating the value for all stakeholders and normalizing it as a practice.
Franko added that the roundtable could highlight the benefits of open behaviors, providing evidence against the “perceived” negatives. Standardizing the language for open practices and incentives could also support their broader use, said Konforti.
Jerry Sheehan of the National Institutes of Health raised the need for new approaches to accommodate a broader set of research outputs, includ-
ing negative results and metrics, into assessments. He added that policies to shorten embargo periods and promote artificial intelligence-ready datasets could serve as motivators for data sharing and access, removing these barriers.
Katie Steen of the Association of American Universities advocated for further developing an infrastructure to support open scholarship systems. This could include, for example, conducting mapping and scoping exercises with an eye toward implementation and developing tools and methods for citing various research outputs that translate to metrics.
Adam Jones of the Gordon and Betty Moore Foundation discussed the need for credit and tools to support the use of open science related to personal persistent identifiers, capturing data downloads, and dataset usage.
(4) Build upon policies and processes already in place to support open science. Several breakout session participants encouraged the roundtable to lean into the building blocks that already exist in open science—for example, elevating success stories, supporting faculty champions, leveraging the role of libraries, and broadly disseminating toolkits. As Fitzpatrick and Farley noted, the roundtable can expand on existing efforts to share best practices to further promote open science.
Participants discussed the importance of identifying open science champions and disseminating success stories about their work. As Poldrack noted, the roundtable can work to establish more influential and prominent ways to highlight the work of open science champions, for example, through “Dean’s Lists” and awards.
Heather Joseph of the Scholarly Publishing and Academic Resources Coalition proposed ways the roundtable can further build a coalition of the willing by challenging funders to create new programs to underwrite a diverse set of institutions to run open institution pilots. This could be supported with a network of university mentors to socialize successes.
Some participants emphasized the role of libraries in advocating for open science. Poldrack said the roundtable could raise the profile of library resources and share information about the potential for collaboration to support and mentor open science practices.
Hanisch recommended making success stories more visible, including those that address major challenges for recognition of open science and data sharing, especially discipline-dependent examples both within and outside the physical sciences. Franko also discussed the need to demonstrate the impact of open science with articles that showcase how expanding access to research can broaden knowledge.
Participants reflected on the potential of working with professional societies to help change norms in open science. Anderson suggested the roundtable systematically identify people who are willing and able to serve as champions for open research and scholarship in different professional societies. Broadly disseminating promising practices that have been adopted by professional societies, such as the Linguistic Society of America, American Geophysical Union, and Council of Graduate Departments of Psychology, among others, could encourage others to follow suit. Bourg agreed, adding that the roundtable should work closely with societies to move to open and overcome real and perceived barriers in this area. This could include, for example, partnering with societies to help their members share other kinds of open science research and outputs such as data, code, and publications.
Several participants discussed broader dissemination of the roundtable’s toolkit and other resources to promote open science. For example, Konforti said that the roundtable should highlight how the toolkit is being leveraged at other institutions to incentivize change at universities that are not yet on board. Franko noted that the “Sending Signals Rubric” section of the toolkit can be used to intentionally engage university departments in an assessment of their level of engagement with open science. It can also be used to encourage institutional leadership to articulate where they stand on open science practices and address gaps between policy and on-the-ground practices at the institutional level. The tool could serve as a scoring system to guide institutions in a move toward open science practices.
Participants described the need to develop other tools and templates for promoting open science. Sheehan noted that the roundtable should support and provide guidance, models, and examples about open science practices including the use of repositories. Joseph discussed the need for template language for faculty senate resolutions and statements on institutional support for open science as well as fact sheets and data on the benefits of open science.
(5) Identify and address the financial and human costs of open science. Many participants commented that the roundtable should help to identify and evaluate the financial and human costs of open science honestly and proactively. To be successful in changing the culture, there should be an acknowledgment of the real costs of open science, who is burdened by those costs, and conversation about what action can be taken to address these costs at the university and discipline levels. Brooks Hanson of the American Geophysical Union and others discussed the importance of investing in
capacity, infrastructure, and people to ensure that open science is done well. Dougherty highlighted another specific cost—data archiving—as a major issue, especially in fields with large datasets. There is also a need for concrete financial support for open leaders through salaries, additional capacity, and funding, as stated by Gibson and other participants.
However, as Konforti noted, despite these initial costs, the larger benefits of open science, such as contributing to research on social justice issues and accelerating scientific discovery, should be communicated more broadly.
