A preprint is traditionally an early version of a scholarly article that is published online before being published in a journal. These days, however, a preprint may or may not end up being submitted to a journal for publication.
The preprint may not have been peer reviewed, although there are now new ways of managing preprints where this does happen.
The preprint process involves uploading your work to a preprint server, where it is publicly available for comment and feedback.
The preprint can be the same as the version of the work submitted to a journal.
This video features members of the UK Reproducibility Network community sharing their experiences of preprints.
Disseminating your work as a preprint benefits you and your readers.
Using a publicly available preprint service gives you publicly time stamped certification and registration of your work. This can help you avoid being ‘scooped’ and can also help identify instances of plagiarism.
Preprints on some servers are open to comments and feedback. This can help you improve your manuscript and highlight potential collaboration.
Rapid dissemination of research findings addresses the problem of delays between submitting to a journal and actual publication.
Your work is accessible to those who cannot afford subscriptions or who are outside academia, such as businesses, charities, policy makers and the general public.
Open access items get high visibility. This can help increase your readership and citation. Some servers give you the numbers of hits and downloads per item.
There is no cost to the author or the reader.
It is possible to track the changes to a discussion because the server retains previous versions.
You usually retain the copyright of your work on preprint servers.
Preprint servers usually assign a unique identifier to each item, enabling your work to be easily cited.
Some funding bodies encourage, or even require, you to use preprint servers.
It is increasingly common in some disciplines for a preprint service to be the ‘go to’ location to find out about new work in a specific field of study.
When your work is published, some preprint servers automatically add a link from the preprint to the published version (as long as the titles match).
On some preprint servers, you can submit direct from the preprint server to some publishers’ article submission systems (for example PLOS journals).
There are some things to be aware of before you decide on preprint. You will need to weigh up these considerations against the benefits we’ve already seen.
Although most journals will accept manuscripts that have previously been disseminated on preprint servers (usually if they are not-for-profit), some do consider preprints as prior publication. You can check journal details using Sherpa Romeo. Bear in mind, journal policies are generally changing on this point. You may decide to select a different journal.
Some authors are concerned that early versions of the work will be read and cited rather than the final version, which could be confusing and inaccurate.
Some journals require manuscripts to be anonymised for double blind peer review. If a preprint version of the work is available, reviewers are able to discover the identity of the author(s). This can be balanced against the benefits of open commenting. The journal’s policy on deposit of preprints can be contradicted by its use of double-blind peer review.
Some disciplines, for example medical sciences, are sensitive to findings being released before they have been peer reviewed, because they might be misinterpreted by non-specialist readers or the press.
Include as much descriptive information (metadata) about your paper as possible to help people discover it.
Upload a copy of your paper. You may also be able to upload accompanying data or other supplementary materials.
A preprint server (also known as a preprint repository) is an online service that allows authors to upload, describe and disseminate preprints.
Most servers make the work freely available and accessible to anyone.
The servers do not usually charge fees to authors or readers for using their service. They may be supported by an institution, such as Arxiv. A few are owned and operated by commercial companies, for example the Social Science Research Network (SSRN).
Preprints, copyright and licences
The author usually owns the copyright of the preprint. If you are the copyright owner you can assign a licence to your preprint to describe how people may use your paper. Many authors choose to use one of the Creative Commons licences. The choice of licence is up to you as copyright holder.
The preprint can be updated to include changes made by the author, for example as a result of open or closed commenting. In the rare instance where an article has to be retracted, the preprint should also be shown as retracted.
Preprint servers usually hold articles that have not been peer reviewed, but this is changing.
Initiatives like the ones listed below allow scholarly discussion on preprint works that can help you improve your paper.
In these cases, reviewers evaluate preprints based on rigorous peer review. They may then ‘recommend’ preprints, which will make them complete, reliable and citable without the need to publish in traditional journals.
If your preprint article is ‘recommended’ in this way, you don’t even need to submit to a classic journal as well – although that is up to you. Recommended articles can still be referred to by researchers and cited in the literature.
Peer Community in (PCI)
An initiative where preprints are recommended following peer review.
In order to be eligible for REF, you have to deposit the final version of published journal articles in an open access repository.
Preprints are officially eligible as long as they are the 'accepted for publication' text, which re-categorises the preprint as 'published' for REF purposes. However, some preprint servers do not currently use the correct metadata for the University to be able to find and report your paper for REF. This means that even if you deposit your accepted manuscript in a preprint server, you will still need to deposit in Oxford University Research Archive (ORA) as well.