Does open access strengthen the link between online attention and citations?
This study finds that while open access strengthens the correlation between online attention and future citations—particularly for hybrid OA articles where downloads drive the link—altmetric scores remain a poor substitute for citations in research assessment and show no citation advantage in high-impact journals like Nature.
Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer
In the world of academic research, the traditional way to measure a scientist's success is by counting how many times their work is cited in other papers. It is a slow process, like waiting for a letter to arrive in the mail; it can take years for a new idea to be read, understood, and referenced by colleagues. In recent decades, a faster signal has emerged: online attention. This includes news stories, blog posts, and social media mentions that happen almost immediately after a paper is published. Researchers call these "altmetrics," hoping they might serve as an early warning system for which papers will become important later. However, a major question has lingered: does this online buzz actually predict future citations, or is it just noise? The answer might depend on who can read the paper. If a paper is locked behind a paywall, only a few people can see it, but if it is free for anyone to read, the connection between online buzz and future academic impact might change.
Two researchers, Ela Ankaralı and Mustafa Mert Ankaralı, set out to investigate this relationship by looking at a massive collection of scientific articles. They focused on a specific type of journal called a "hybrid" journal, which is a publication that usually charges a fee to read but also allows authors to pay a fee to make their specific article free for everyone. This setup created a perfect natural laboratory. Because the articles were published in the same journals, often by the same editors and in the same fields, the researchers could compare free articles directly against paid ones without the confusion of comparing different types of magazines. They examined more than 60,000 articles published in 2018 across 1,275 different journals. They tracked the online attention these papers received in 2021 and then watched to see how many citations they accumulated between 2022 and 2026.
The researchers found that online attention does predict future citations, but the strength of that prediction depends entirely on whether the paper is free to read. For articles that were locked behind a paywall, the link between online buzz and future citations was weak. A paper might get a lot of tweets or news mentions, but that did not strongly suggest it would be cited by other scientists later. However, for the articles that were made free at the moment of publication, the link was much stronger. When a paper was open for anyone to read, the online attention it received was a much better indicator that it would go on to be cited. The data showed that for free articles, the correlation between attention and citations was nearly double that of paid articles.
To understand why this happens, the team looked at what people actually did after seeing the online mentions. They discovered that the key step was downloading the paper. When an article was free, online attention led to many more downloads. These downloads, in turn, were the strongest predictor of future citations. The logic is straightforward: if a paper is free, the people talking about it on social media or in the news are more likely to be able to read it, and if they read it, they are more likely to cite it later. If the paper is locked behind a paywall, the online attention might generate curiosity, but that curiosity often hits a dead end because the reader cannot access the full text. The researchers also found that this effect was driven mostly by public attention, such as news and blogs, rather than by the activity of professional researchers saving papers in their personal libraries.
The study also uncovered a significant inequality in who benefits from this system. The ability to pay the fee to make a paper free was not distributed evenly. Authors from wealthy countries and top-ranked universities were far more likely to have their work made open access. Consequently, the boost in citations and attention that comes from being free was concentrated among these well-resourced groups. To test if this was truly a cause-and-effect relationship rather than just a coincidence, the researchers looked at a "natural experiment" involving national agreements. In five European countries, governments negotiated deals that automatically made articles from their researchers free, removing the need for individual authors to pay. In these countries, the researchers found that articles became free much more often, received twice as much online attention, and were downloaded three times as often. Most importantly, these articles received 29 percent more citations than similar articles from other European countries that did not have such agreements. This suggests that making research free genuinely increases its impact, rather than just reflecting the quality of the work.
However, the researchers also found limits to this advantage. In the most prestigious, high-impact journals, where almost every university already has a subscription, making an article free did not provide a significant boost in citations. The benefit of being free was most pronounced in specialized journals where fewer people have access. Furthermore, while online attention is a better signal for free articles, it is still a poor substitute for the traditional citation count when trying to predict which papers will be the absolute best. Early citations remained a much stronger predictor of future success than any online metric. The study concludes that while online attention is a useful tool for spotting interesting work, its value as a signal depends heavily on whether the work is accessible to the people talking about it. If research is free, the conversation online is more likely to turn into real academic impact.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.