Giusy is throwing some deep dives into the trenches of scientific literature. Deep they are indeed. Almost running out of oxygen whilst frantically flapping through the murky waters near the surface, she comes across a not-yet-published paper (or better: a non-peer-reviewed paper) that’s floating around the internet. Apparently well-written and soundly researched, she goes looking for more and retrieves some gems from a single location. Yet there seem to be more of those, so-called preprint servers. Giusy wonders:
Which preprint server do you recommend?
Dear Giusy,
None of them. Go back and get those scuba goggles back on your face. Fill the oxygen tanks and submerge yourself back into those murky waters you just came out of. Just go a bit deeper. A teensy-weensie. No, a whole lot more! Unwilling maybe, but you’ll be way happier afterwards. I promise. The thing is; as a scientist you’re kinda dependent on the integrity of other scientists. Mostly in terms of the validity of research (and reproducibility, clarity in description, honesty about all of the data/results presented, bla bla…), and a little in terms of being a colleague at the same institute (honest work reviews, no teasing, proper recommendation letter writing, bla bla …). Peer-review being a pillar of the validity of research. I mean, you wouldn’t expect researchers to just throw some incomplete work out there willy-nilly, would you? It would only end up online after quadruple-checking the data, methodology, results, interpretation and discussion, right? It’s a good thing you have that much confidence in others. As a personality indicator, I mean. Not so much as a scientist, unfortunately. An unrivaled level of cynicism is key for those quirky dwellers of the deep seas. You do realize that the number of rejected papers for journals are through the roof (and increasing each and every year, just as the number of retractions by the way). Do you know the rejection rate of preprint servers? No, I don’t mean because someone didn’t fill in the forms completely correctly, or plagiarism, … . No, I’m talking about actual rejection because of the contents. Now, do you know? They don’t reject as long as you seem to provide a scientific paper. This includes wild claims, incomplete analyses, circular logic, deceitful and/or incomplete methodology description, mistakes, unbalanced reviews of the literature, … and the list keeps going … and going … and going on. In short, anything you come across at a preprint server (whichever one you’re searching through) should be critically assessed. By you. Way more critically compared to those papers published in peer-review journals. What if you base your entire 12-month long study on some ‘data’ you came across in a preprint document, supposedly submitted to an actual journal, after which you find out that the paper was never published (anywhere) and the document was ‘removed by authors’ from the preprint server? What do you cite? Is your study even valid?
Sure, you’ll come across some good ones on a preprint server. Don’t be surprised that there are some scientists out there that are actually doing a proper job. Yet it is also bulging with not-really-that-great documents. I doubt there’s one reader of this blog post that has submitted a paper and gotten it published without a single comment from a reviewer/editor that helped clarify an issue, elucidate the methodology, balance the citations, …. et cetera. That’s the thing with scientists; we all think we’ve written a gem, yet it turns out to be just a polished piece of glass.
Yours truly,
The Sassy Scientist
PS: This post was written and posted when it was completely finished. Like it should be.