PLOS Note: we use this blog, on occasion, to highlight authors and their research. Today, we are shining a spotlight on this work that was published last week on PLOS ONE. The topic is dear to our heart. The author of the blog and the paper is Marc-Andre Simard. He is a Ph.D. student in information science at the University of Montréal who works on open science and science policy.
The COVID-19 pandemic has heightened the importance of faster and more efficient dissemination of scholarly literature. Early in the pandemic, several publishers, such as Springer-Nature, the American Association for the Advancement of Science (AAAS) and Elsevier, announced the opening of their COVID-19-related papers. However, it remains unclear if these papers will remain free to access for researchers and the public. For instance, Elsevier mentioned that their Novel Coronavirus Information Center will be available “as long as necessary,” hinting that these resources might be locked behind a paywall one day.
The movement toward open access
The scientific community has long advocated free and unrestricted access to scholarly literature. This movement toward open access was formalized through 2002’s Budapest Open Access Initiative. The idea behind the open access (OA) movement is that science is a public good that should be made available to everyone in order to remove some of the technological and financial barriers to science and accelerate research and education across the planet. Over the past 20 years, several major initiatives have been developed worldwide to support open access dissemination. These include UNESCO’s Recommendation on Open Science and Plan S, a European initiative that aims to make research articles supported by public funds freely available online, without embargo, and with transparent pricing.
Just how big is open access?
Using the Web of Science, a database that indexes publication and citation data from about 13,000 journals annually, and Unpaywall, an open database tracking open access content from over 50,000 scientific publishers and repositories, we have collected a dataset of 8,137,675 documents that were published between 2015 and 2019. We found that over 40% of all scientific publications indexed in the Web of Science during that period were available in OA at the time of our data collection, with shares ranging from 21% in humanities to 50% in medical and health sciences (Figure 1).
Figure 1. Share of open access papers by discipline.
Does open access really matter for lower income countries?
Scientific publishing is a lucrative business. Since the early 2000s, more than half of scientific papers have been published by the so-called “oligopoly of publishers” that have been making record profits. Participating in science is also expensive: in order to have access to the latest scientific discoveries, it is often necessary to pay journal subscription fees that are so high that even some of the wealthiest universities in the world have decided to unsubscribe from them. Low-income and developing countries are particularly vulnerable to these pricing dynamics.
Using the World Bank’s country classification by income and author affiliation data from the Web of Science, we have looked into the relationship between income level and 1) open access publishing and 2) references made to open access literature. Figures 2 and 3 illustrate OA publications and references of countries compared to the world average. Our data show that low-income and lower-middle-income countries, especially in Sub-Saharan Africa, tend to publish and use open access literature more than the rest of the world. However, upper-middle-income countries use open access literature less than the rest of the world, while open access practices vary greatly in high-income countries.
Figure 2. Normalized map of the number of open access publications per country. Green indicates that a country is over the world average, and brown indicates that a country is under the world average.
Figure 3. Normalized map of the number of references made to open access per country. Green indicates that a country is over the world average, and brown indicates that a country is under the world average.
Those findings show that countries in sub-Saharan Africa publish and cite open access literature at a higher rate than the rest of the world. At the same time, we found that the Middle East and Asia are the areas where both their proportion of OA publications and of OA use is lower. This could be explained by the fact that most publishers provide publication fee waivers for low-income countries but not or only partially for middle-income countries. Ultimately, our study highlights national differences in open access uptake and leads us to believe that more open access initiatives at the institutional, national, and international levels are needed to support the broader adoption of open scholarship.