Preprints: best practice tips librarians can share with researchers
15 December 2021 | 8 min read
By Jay Bhatt, Linda Willems
In a recent university presentation on preprints, Engineering Librarian Jay Bhatt and his colleagues highlighted some of the key points to consider
In the image above: Outside Hagerty Library, Drexel University Libraries - Jaci Downs Photography One of the most striking aspects of COVID-19 from a publishing perspective, has been the rapid growth in preprints. Some estimate(opens in new tab/window) that researchers were posting around 40 new virus-related preprints each day at the height of the pandemic.
With this avenue for sharing results on the rise, now seems the ideal time to pick the brain of preprint advocate Jay Bhatt, Librarian for Engineering and Biomedical Engineering at Drexel University(opens in new tab/window), a private research university in Philadelphia, USA.
In this article, Jay runs through some of the key insights he shared during a presentation on preprints(opens in new tab/window) for this year’s Open Access Week(opens in new tab/window). He touches on the potential implications for researchers and research, before highlighting some key dos and don’ts that librarians can share with authors.
A quick guide to preprints
Preprints are typically full draft research papers that are shared publicly prior to peer review. Preprints posted on recognized preprint servers have their own digital object identifier (DOI) and are citable.
These preprint servers are usually green open access repositories, and many are subject specific, for example, Health Sciences’ medRxiv(opens in new tab/window) and Biological Sciences’ bioRxiv(opens in new tab/window). There are also more broad scope preprint servers, such as SSRN(opens in new tab/window).
Support for preprints as a research sharing channel is growing. For example, in July 2020, Europe PMC began indexing the full text of COVID-19-related preprints alongside peer-reviewed articles, to make them searchable. And Elsevier’s Scopus now includes preprints published from 2017 onwards.
Although many journals won’t consider papers previously published elsewhere, manuscripts that have already been shared as preprints are generally acceptable: in return, most publishers ask authors to update any pre-publication versions with a link to the final published article.
Preprints – the pros and cons
When Jay talks with researchers about preprints, the benefits he highlights include:
Speed: “Preprints are a great way to share research quickly and get it discovered early,” he explains. “And because preprints are time stamped, it helps to establish who came up with an idea/solution first.”
New opportunities: Another big plus point for Jay is that “preprints can lead to researchers discovering compatible interests, partnership opportunities, interdisciplinary avenues of research, or even result in joint grant applications. If you look at award eligibility criteria today, funders put a lot of emphasis on collaboration. And having preprints on a platform where multidisciplinary research is available helps collaborators find each other.”
Improved manuscripts: According to Jay, several graduate students he’s spoken with have used the comments posted on their preprints to refine their papers before submitting them to a peer-reviewed journal.
Ability to share negative results: Jay says: “Null or negatives results have a lot of value to the community - significant time and money can be saved by sharing them so others don’t make the same mistakes. In fact, in situations like COVID, sharing negative results can literally save lives.”
According to Jay, other advantages of preprints include the fact that they can be cited, allowing authors to start accumulating citations right away. In addition, they are freely available, which can help to accelerate and improve research, and ensure the work reaches a wider audience, including the public.
But preprints also have their disadvantages, as Jay is quick to admit. These include:
Accountability: The results in preprints haven’t been validated by peer review, yet some members of the public and media quote them as confirmed facts; this has been a particular issue with some COVID preprints. Jay says: “It’s important that academics communicate more clearly with non-academic audiences about the limitations of preprints. However, the controversial COVID preprint studies were quickly disproved and results like that can help to indicate a more responsible way forward. In that sense, controversy can be both good and bad.”
Challenges around tracking impact: Jay says: “Until we can find a way to link the citations that preprints receive with the citations that an associated peer-reviewed article receives, we won’t have a true picture of a paper’s impact.”
For Jay, other challenges include the fact that preprints can make it difficult to maintain the anonymity of double-blind peer review when the paper is submitted to a journal. In addition, if the preprint has received a negative reception, that can influence the perception of the final published paper.
Jay’s top dos and don’ts for researchers posting preprints
In recent years, Jay and his colleagues in the Drexel Libraries have developed a series of tips that they - and other librarians at the university - can share with authors.
DO maintain version control. “It’s important that readers understand whether they are reading version 2 or 3, etc., and there should always be a link back to the original version.”
DO connect your preprint to your ORCID ID – that ensures that your ORCID ID is included in the preprint server’s metadata and any preprint DOI metadata. The preprint is also included in your ORCID ID record. Jay explains: “Essentially, it ensures that the author is connected to their preprint.”
DO check whether any publisher/journal you are considering submitting to is willing to publish preprints. Jay says: “The Sherpa Romeo(opens in new tab/window) website is a great resource – it lists all the publisher policies on preprints.”
DO choose a recognized preprint server. According to Jay: “There are so many good, reputable servers out there. If you aren’t sure about a server, you can ask colleagues for advice, look at the server’s policies, at how many preprints it hosts, and whether there have been any ethics issues with preprints published on that server. If there has been a number of retractions, that should probably raise a red flag.”
DO cite your preprints, where relevant. Jay explains: “Many preprint servers track usage, e.g., downloads, so when you are thinking about where to post, it’s worth looking at the metrics that server provides. These can then be cited in CVs – particularly helpful for students applying for postgrad positions. And funders such as the US National Institutes of Health and the UK’s Wellcome Trust now allow researchers to cite preprints in their grant applications.”
DON’T deposit your preprint in multiple servers. “This not only confuses other researchers, it also makes your impact metrics and reader feedback really difficult to track,” says Jay.
Using preprints in research
With the number of preprints on the rise, and popular sources for literature reviews, such as Scopus, now indexing them, researchers increasingly encounter preprints when conducting their background research. According to Shirley Decker-Lucke, a Director of Elsevier’s preprint server, SSRN, “more and more researchers in every discipline are consulting preprints. At SSRN, for the past two years, we’ve seen preprint usage rise by 30 percent per annum.”
Because SSRN hosts all sorts of early-stage research, including preprints, working papers and conference papers, it can be hard for readers to know which version of a research paper they are viewing. “For this reason, we are in the process of adding ‘This is a preprint’ banners at the top of papers we know to be preprints,” says Shirley. “In the instances in which we know there is a published journal article version, we will also add a notice and a link to the Version of Record. This is a big step towards making it easier for readers to know what they’re looking at and for them to find the most recent, peer-reviewed version of the research, when available.”
Jay also has some advice for researchers who are considering citing a preprint in their research. “Again, it’s important to check that the preprint is hosted on a reputable server and I also recommend reviewing the publisher or journal policy on citing preprints; most journals now allow preprints to be included in the reference list, and organizations like the NIH have developed preprint citation guidelines."
What does the future hold for preprints?
Jay expects we will continue to see the number of preprints grow, with new preprint servers launched to cater for specific research interests, such as the ultrasound community’s new FocUS Archive(opens in new tab/window) service. Jay adds: “There will always be some limitations because preprints are still not peer reviewed. Having said that, I do believe the benefits of preprints far outweigh their limitations.”
Jay also believes more education is necessary to improve the quality of preprints. “As librarians and universities, we need to teach graduate research students and researchers about integrity and ethics so that they make ethical choices. We also need to increase their awareness of preprints – at the moment, that’s lacking.”
In addition, Jay would like to see publishers unite to increase the discoverability of preprints – potentially with a single search repository. And he believes that publishers can also play a role in resolving some of the issues around tracking and linking the citation and usage metrics received by a preprint and the final published paper.