Christie Bahlai believes using open data hurt her chances at her first post-PhD job interview, for a tenure-track faculty position at a large university in the southern US. Bahlai says she got an unexpected reaction from the search committee chair when she explained that her research mostly used other people’s data to analyze insect population dynamics. She recalls him asking her, “What happens when the data runs out?” To her, it seemed like the chair didn’t understand how much open data was available online for people to mine and use in their own research. When she tried to explain it, he seemed skeptical. “He couldn’t wrap his mind around the fact that data was being shared,” she says.
To Bahlai, open science represents the democratization of science, with researchers making their data, code, and other resources used in a study available for anyone to see. “The idea is that the process of science should be available for other scientists to build on and for the public to look at,” she says, noting exceptions such as private medical data that wouldn’t be appropriate (or legal) to share.
As Bahlai’s experience shows, scientists aren’t always rewarded for conducting research in accordance with open science principles. A new initiative plans to change that. The Higher Education Leadership Initiative for Open Scholarship, or HELIOS, which launched this March, is a coalition of more than 75 member colleges and universities that have committed to fostering open science practices, including through their hiring, promotion, and tenure decisions.
A legacy of individualism in science
HELIOS is a collective effort to move away from the competitive model in life sciences research that Keith Yamamoto, the vice chancellor for science policy and strategy at the University of California, San Francisco (UCSF), traces to the 1970s. Yamamoto, who leads strategic discussions among HELIOS member institutions (including UCSF), says that during that decade, Benjamin Lewin, who founded the journal Cell, began attending scholarly presentations and recruiting what he viewed as the best papers for publication in that journal. Over time, publishing in Cell came to be seen as prestigious.
Other journals followed Cell’s lead, Yamamoto says, and they began competing against one another to publish cutting-edge research. They also started denying publication to scientists who’d shared their research publicly, and because scientists badly wanted to see their work in these top-tier journals, they became more hesitant to share and discuss unpublished results.
“The scientists bought into it,” Yamamoto says, adding that he doesn’t blame Lewin for coming up with an innovative marketing strategy. The journals wouldn’t have succeeded in shifting the culture if the scientific community hadn’t bought into the concepts of prestige and status, he says.
Yamamoto says this competition to publish work in prestigious journals led to an emphasis on individual contributions over collaboration in academia today. For example, tenure committees heavily weigh publication as a first or senior author, especially in prestigious journals—a process that can take several years, delaying when others have access to advances in scientific knowledge, he says. Yamamoto says it’s also common for committees to completely disregard papers where the tenure candidate is listed as a middle author.
Those individualistic values aren’t limited to universities and colleges. Grant agencies, for example, may decide to deny funding to a group of researchers if they get scooped by another team investigating a similar problem, Yamamoto says.
“So those kinds of values and practices then serve a very strong disincentive for an investigator to practice open science,” he says.
A vision of open science
Geeta Swamy, who is both the vice dean for scientific integrity at Duke University and the strategic lead for HELIOS, writes in an email to The Scientist that sharing data, publishing in open access journals, and posting research findings as preprints make research more inclusive, transparent, and efficient.
“To tackle big societal challenges like climate change or COVID, to better understand our place in the universe, to build a more equitable and just society, we need to engage with more communities, include more people from more corners in the research conversation, and allow our work to be accessed, tested, and built upon more rapidly,” she says.
Bahlai, now at Kent State University (which is not a HELIOS member), similarly argues that open science practices support her institution’s mission to further scientific discovery. “If I developed a method that many other ecologists were using to do data analysis, that would potentially have more influence in scientific fields than just a paper that shows me using it,” she says.
HELIOS wants to bend academia’s incentive structures toward cultivating collaboration. To accomplish this, like-minded institutions have gathered several times since 2021—beginning with a roundtable discussion convened by the National Academy of Sciences—to discuss priorities and strategies. The proceedings of a 2021 member workshop, “Developing a Toolkit for Open Science Practices,” includes language that institutions can use to show students and faculty their commitment to open science. The toolkit also includes templates for evaluating open science practices in job and tenure applications with example criteria including publishing in open-access journals, posting data using FAIR (findable, accessible, interoperable, and reusable) principles, and sharing other research outputs such as computer code.
Yamamoto suggests that journals also have a role to play in incentivizing open research, such as by handling authorship differently. Instead of having first, middle, and last authors, he says papers should have an alphabetical list along with a description of how each contributed to the work. This approach would highlight to tenure review committees what’s really important, he argues. “It simply is a reminder to the tenure promotion committee that the journal actually doesn’t matter—it’s what’s in the paper, and that the order of authors doesn’t matter—it’s what the author did.”
Yamamoto also says tenure committees should give credit to scientists who give their time to peer reviewing papers and other research outputs, which they currently do not. In addition, he argues that helping to educate the public and building trust in science is something tenure committees ought to consider.
Lara Mangravite, president of Sage Bionetworks and board member for the Center for Open Sciences, is not involved in HELIOS, but says that in her view, it’s on the right track. “I think the only way that open science practices are going to get broadly adopted is if they are integrated into those activities: tenure, hiring, promotions,” she says.
Mangravite notes that there are already signs of a cultural shift outside of universities themselves. Many grant agencies, such as the National Institutes of Health, now require projects they fund to put their work in the public domain, she says, which is pushing journals to become more open as well. “And now it’s time to bring the [academic] institutions in,” she says.
Yamamoto notes that it’s a struggle to change the culture across an entire institution, let alone all of academia. “To get everybody to fall in line at the same time—everybody recognizes this is not possible,” he says. When people begin to realize the magnitude of shifting required, he says many throw their hands up in dismay. The idea is for HELIOS members to join forces and begin to enact change together.
Some schools and departments within HELIOS member institutions are already encouraging open science behaviors as part of tenure and promotions. Swamy says Duke University highly values open science; she provides excerpts from its School of Medicine’s guidelines, which state that with respect to promotion and tenure decisions, participating in open science will be considered “under the broad category of service” to the school. She also points to updated criteria for promotions within the University of Maryland’s Department of Psychology that recognize the importance of publishing datasets, research tools, and software, as well as from the University of Virginia’s School of Data Sciences, which says “tenure requires a commitment to open science,” such as publishing in open access journals, collaborating, and publishing data and software.
I generally believe that, first of all, science is something that’s done in the public good.
—Christie Bahlai, Kent State University
Despite incremental moves toward linking tenure decisions to good opening science practices, a major obstacle remains: senior faculty members who often hold leadership positions and are resistant to changing the system that they succeeded in, Yamamoto says. In contrast, junior faculty—many of whom saw how well collaboration worked in fast-tracking research on SARS-CoV-2—generally like collaborating and support open science, he says. The result is two factions with opposing ideals. Yamamoto projects the culture will shift as the older generation retires and younger faculty gravitate toward institutions that clearly value open science practices.
Mangravite says she “one hundred percent” sees this divide between senior and junior faculty. But she says that rather than waiting for older faculty to retire, what’s needed is to incentivize younger faculty to participate in open science now instead of continuing to hold them to traditional standards set by more senior academics.
Take Bahlai, for example, who worked as a postdoc for three years after her ill-fated faculty interview. Then, in 2017, she landed an academic position at Kent State, where she’s currently a computational ecologist in the Department of Biological Sciences. Now she’s up for tenure, but she’s struggling to figure out how to incorporate her open science practices into her application.
“I generally believe that, first of all, science is something that’s done in the public good,” she says. “It’s a collectivist endeavor.” However, she says it’s time-consuming to share data, code, and teaching materials while still achieving Kent State’s more traditional benchmarks, such as publishing two papers per year.
“I think that’s exactly the problem, right?” says Mangravite, regarding young faculty in situations like Bahlai’s. Why should junior faculty prioritize time to open science when people reviewing tenure applications may not pay attention, she asks. “I think that’s why this initiative is so important.”