How Open Science Can Both Advance and Hinder Equity in Research
In January 2023, the White House Office of Science and Technology Policy and the National Science and Technology Council released an official definition of open science for use by the US government: “The principle and practice of making research products and processes available to all, while respecting diverse cultures, maintaining security and privacy, and fostering collaborations, reproducibility, and equity.”
While this definition promises equity, not all aspects of open science currently deliver.
(New to the concept of open science and what it comprises? Check out our primer.)
“In general, open science seems like this beautiful thing,” says Verónica Andreo, whom NC State University’s Center for Geospatial Analytics hosted as a visiting scholar this past spring. Andreo is a researcher with Argentina’s National Council for Scientific and Technical Research (CONICET), a lecturer for the National Commission for Space Activities’ (CONAE) and National University of Córdoba’s Gulich Institute, and chair of the GRASS Project Steering Committee, which oversees development of the open-source geospatial modeling engine GRASS GIS.
Andreo’s experience as a scientist in Argentina mirrors the experience of scientists across the world: Open-source software can help level the playing field for researchers who are at an economic disadvantage, she says, but the costs of open access publishing can exacerbate existing inequities.
Open-Source Software Levels the Playing Field
Geospatial analysis, like statistical analysis, requires specialized computer software to accommodate handling and processing large amounts of data. For many scientists, though, the licenses of commercial software are prohibitively expensive.
“From my standpoint,” Andreo says, “as someone coming from the Global South, open source is indispensable. We cannot afford software licenses. Open source allows us to do our work. It is a key component of our work.”
As a student in Argentina, Andreo used the freely available open-source statistical software R in lieu of costly proprietary packages but was limited to using proprietary GIS software. “There must be something equivalent in GIS that is open source,” she recalls thinking. After learning about GRASS GIS, she met the GRASS community in 2013 and became its elected head seven years later.
In 2013, Andreo knew she wanted to not only use GRASS GIS but also support it. At the time, she didn’t yet have expertise in writing code, she says, so she worked on GRASS tutorials and manuals, providing examples of how code could be used and translating documentation into different languages. Soon she could write small pieces of code, then her own extensions for the GRASS GIS software itself.
“[GRASS] is a really welcoming family and community,” Andreo says, “and that’s why I stayed [involved].”
A new $1.5 million grant from the National Science Foundation, led by researchers at the Center for Geospatial Analytics, will expand the GRASS community over the next two years, expanding access to and support for this freely available geospatial software. “It will support facetime to grow the community and sustainability of GRASS GIS when we’re no longer here,” Andreo says.
Use of the open-source software is free, but that means the people who develop and maintain it typically have not been paid for their efforts. The new funding will help modernize software infrastructure and strategically grow the GRASS community to achieve a technologically and socially sustainable open-source ecosystem.
In all of her current work, Andreo advocates for the development and use of open-source solutions, for both teaching and research. This approach increases access to important research tools for students and professionals alike, regardless of their funding level, advancing equity.
On the other hand, she notes, the costs of open access publishing can create problematic, self-perpetuating disadvantages.
Open Access Shifts the Burden of Cost
In science, a research project is considered incomplete until it is published––that is, until its findings are reported in a peer-reviewed academic journal. Traditionally, researchers paid low-cost page charges or, more typically, nothing at all for publishers to distribute their work, and institutional libraries or individual readers paid subscription fees to access scientific publications.
Open access publishing, one component of open science, disrupts this model by making peer-reviewed papers free to read and reuse…but very expensive for scientists to publish.
“As a researcher in the Global South,” Andreo says, “I cannot publish open access. We cannot afford the article processing fees.” Publishing just one open access article can cost a researcher thousands of dollars.
The high price contributes to an unfortunate cycle, Andreo explains, that disadvantages scientists outside of the wealthiest nations as well as scientists in wealthy nations who lack resources: Well-funded researchers who can pay for open-access publishing typically reach a wider readership and are therefore more likely to be cited by others. High citation rates are rewarded in science, including by funding agencies, leading to more grants being given to the already well-funded (and highly cited) scientists.
“It’s a positive feedback loop for the privileged,” Andreo says.
Some open-access publishers have created systems that provide publishing discounts to scientists who review submissions for them, but these can be considered predatory, Andreo notes. Other publishers have created agreements with research institutions (including universities like NC State) to reduce the publishing costs for individual affiliated scientists. When it comes to open access, “It’s a complex ecosystem,” Andreo says. “It’s not as simple as everyone has access. It has a cost.”
Open Science and the Academic Rewards System
The monetary cost of open-access publishing and the unpaid labor of open-source development can disadvantage underprivileged scientists, and so Andreo emphasizes the necessity of rewarding other forms of research effort and scientific contributions.
A researcher’s productivity is typically measured in peer-reviewed papers published and grant funding secured, and these criteria determine career advancement in academia. Yet, the creation and curation of other products, like open datasets and open-source software, are also important to the scientific enterprise.
There is interest in changing the rewards system, Andreo says, so that other products or outcomes of research beyond publications are valued and rewarded. But, she explains, the change cannot come from the most vulnerable people in the system, that is, the early career researcher who is trying to establish themselves within the bounds of the old rewards system.
“Senior scientists should start the change,” Andreo says, “because they are in a comfortable position. They have nothing to lose, and they have all the power.” Researchers who have already reached the highest echelons of science by publishing frequent and highly cited papers throughout their careers “won’t be fired because they are not publishing [now],” she explains. To advance equity for the researchers coming up after them, senior scientists “should value products of research” like data and software, Andreo says, because these scientists have considerable influence over how scientific productivity is measured and what is valued, including efforts to advance open science.