Preprint is a term used for academic articles that have not yet undergone peer review. Typically, it is the same version of text that is submitted to a journal for peer review.
Preprints are typically shared for two reasons:
to speed up the sharing of results with the scientific community
to ensure access to research findings
The peer-review process can take a very long time in some disciplines, and speeding up the exchange of results can be important particularly when researching topical issues. You can therefore share a preprint at the same time as submitting the article to a journal for peer review, so that the results reach the scientific community as soon as possible.
Preprints can also be shared at a later stage, even after the article itself has been published. Publishing a preprint at a later stage is done primarily to ensure the availability of research results – for example, a publisher may prohibit the publication of the final version of the manuscript in a repository, unless it is published as open access, or impose an unreasonably long embargo. In such cases, it is possible to make the research results available at least in the form of a preprint and add a link to the final published version of the article.
There are many subject specific platforms dedicated to sharing preprints. The most commonly known are:
arXiv – mathematics, physics, computer science, statistics etc.
bioRxiv – biology
medRxiv – medicine
SocArXiv – sociology
PsyArXiv – psychology
ChemRxiv – chemistry
Although the articles on preprint servers are not peer reviewed, many platforms perform a basic screening, ensuring that uploaded articles are scientific and fit within the scope of the platform.
Preprints can also be shared via multidisciplinary platforms like OSF Preprints, where you can also find a list of subject- or language-specific preprint servers associated with the OSF platform. Another multidisciplinary preprint platform is a server Preprints.org, which is owned by a commercial publisher MDPI, so there are concerns regarding its future development and sustainability.
Preprints can also be shared via generic repositories such as Zenodo or the Charles University Research Publications Repository.
Regardless of where you share your preprints, it is always necessary to clearly state that this is a version of the article that has not yet undergone peer-review.
Some research funders, such as the Ministry of Education, Youth and Sports in its OP JAK calls, classify preprints as a means of early sharing of research results; therefore, when reporting preprints, they require that they be published before the final version of the article is released.
Some journals may have issues with preprints being shared at the same time as the article is submitted. For example, if a journal operates a double-blind peer review, where reviewers and authors do not know each other’s identities, the journal may decide to reject a manuscript with a published preprint, as it would be unable to ensure equal conditions in the peer review process. If you are unsure whether your chosen journal supports the sharing of preprints, check the journal’s website or send an enquiry to the editorial office.
Check that your chosen platform accepts preprints in the format you require; for example, medRxiv only accepts preprints that have not yet been published in a journal (see the FAQ).
Some databases, like Google Scholar, include preprint citations to the total number of citations of a given output. Others, like the commercial Web of Science or Scopus, exclude preprint citations from their official metrics like the Journal Impact Factor. Web of Science, for example, introduced a separate index for preprints called Preprint Citation Index.
Research assessment, however, has been gradually turning away from using solely bibliometrics for assessment and rather reflects the quality of results through commented nominations of selected results or their societal relevance (including the Methodology for the Evaluation of Research Organisations 2025+).
If you are citing a preprint in your work, the citation should include information that the cited work is a preprint. The citation may look, for example, like this:
Surname, N. (Year). Title of the preprint [Preprint]. Preprint Server Name. DOI or URL.
If there is a version of the article after peer review and you have access to it, you should use and cite the final, peer-reviewed article.
Discussions about preprints sometimes include doubts regarding the reliability of results published in preprints. A study published in July 2026 as a preprint looked at how much scientific claims published in preprints on bioRxiv differ from those that are eventually published in journals. The study reports that primary claims remain unchanged (39.9 %) or minorly revised (50 %). The study, of course, only looked at preprints that were eventually published but the authors add that roughly two thirds of preprints published on bioRxiv are eventually published in peer-reviewed journals. Preprints, therefore, appear to be a reliable source of information, however, it is always necessary to keep in mind that the papers have not yet been reviewed.
Residency, Invoicing and Correspondence Address
Charles University
Central Library
Ovocný trh 560/5
116 36 Prague 1
Czech Republic
Office Address
José Martího 2 (2nd floor)
160 00 Prague 6
Phone: +420 224 491 839, 172
E-mail: openscience@cuni.cz
Www: openscience.cuni.cz
