CONNECTIONS: The VDOT Library E-Newsletter

Learning to Read the Fine (Pre)print!

There are pros, cons and limitations to submitting "preprints" to sources like engrXiv.org ("Engineering Archive").

Increasingly, we librarian types are being asked to help our customers understand and assess the validity of research papers found online that do not appear to have undergone (at least yet) the typical peer-review and publishing process used in scholarly communications.

What we are witnessing is the rapid growth of research results that is increasingly being made accessible by authors before (or during) the traditional peer-review process, and the rise of "preprint servers" for making those early draft versions of papers accessible.

Is this the end of the world? No, far from it. In fact it may be great news for engineers and other researchers who want to “connect” through legitimate scholarly communication, but who don’t want to wait on peer-review bottlenecks to initiate their inquiry. That’s where it helps to define and understand what preprints actually are, talk a bit about "postprints" and draw distinctions between those and the publisher’s final "published" version.

Preprint: The author’s early draft of a paper that has been submitted for consideration to a scholarly journal. If posted to a preprint server, that preprint does not reflect revisions that will eventually be made during the peer-review process. The core value of this draft version is the ideas and research methodology articulated in the draft.

Postprint: This is like as the author’s “final draft” of the paper, submitted for publication. The postprint contains all or most of the suggested revisions made by reviewers during the peer-review process. It does not, however, include layout or copy editing (or perhaps things like final graphs and charts)…all things that are done by the publisher’s editorial staff prior to publication. The core value is (presumably) fully articulated scholarship, a more fully articulated methodology and or more complete research data.

Published: This is the version you seen “in print” or the final version of the article produced by the publisher. When dealing with hard-copy publications, this is the printed version found in proceedings and journals. In the digital environment, the published version is usually a PDF or HTML version available through the publisher’s Web site or article databases, and usually has a Digital Object Identifier (DOI). 

Information experts Jay Bhatt, an engineering librarian from Drexel University, point to drawbacks with the traditional academic publishing model, including: slow review process (sometimes taking as long as a year!) and delayed revelation of research findings for researchers trying to "improve" on that research. 

Conversely, Bhatt notes as a benefit the opportunity for researchers to share “null” findings. Turns out, null results are pretty common in the research world. Yet they aren’t likely to get published. On the surface that makes sense, but in science, such research “failures” are an integral part of exploration. After all, finding new ways to share “what not to do” can save researchers and scientists from repeating each other’s mistakes needlessly right?

Well that’s one of many factors driving the creation of Open Access Archives and “preprint archives” like engrXiv.org (aka "Engineering Archive"), which is mentioned along with others in this Library FAQ on preprint servers and Open Archives. 

The unprecedented trove of COVID-19 research that has been produced since the start of the ongoing pandemic is an interesting case in point. Sometimes it takes an emergency to underscore the value of new models in scholarly communications, even while we rely on the time-honored practice of peer review to vet research findings.

A study on publication patterns published in a December 2020 article by Nature, suggested that as many as 200,000 coronavirus articles had been published that year alone. These startling graphics depict what Nature called “an unprecedented flood of research on coronavirus swept websites and journals this year” of which more than 30,000 were “preprints” according to the Dimensions database.

More recently, a press release from Ohio State University points to research that found both exponential growth in the quantity of papers coronavirus published, while simultaneously noted decreases in research partnerships and collaborations among researchers. From January to October 2021, “more than 87,000 papers about the new coronavirus were published worldwide.” Caroline Wagner, co-author of an article that also appeared in Scientometrics noted in a university news release that trend may be unprecedented in the history of science. 

While preprints are a real phenomenon, that doesn't mean all is perfect in the world of preprints...or the world of research in general. 

The biggest downside of preprints? Probably the lack of accountability for findings. Why? Probably because they have not been adequate scrutinized by experts or fully "validated" by peers.  So perhaps there's no free lunch.

Ken Winter
(434) 962-8979
VDOT Research Library