Chasing Paper

December 19, 2014 9:01 AM

An investigation for Scientific American by MeFi's own cgs06 uncovers evidence of widespread fraud in scientific publishing's peer review system. Alarming signs point to the Chinese government as a source of institutional support and funding for questionable papers and fake peer reviewers.

More details on Retraction Watch.

The Committee on Publication Ethics issued a statement today on this issue.

posted by overeducated_alligator (26 comments total)

Just a clarification from reading the Scientific American piece - what is the view of having your research written up by an outside source? You legitmately do the research, you make the conclusions, but your communications skills or grammar are lousy, so you hire a third party to write the paper. As long as you acknowledge the third party as an author, even if they did none of the bench or whiteboard, is that OK?

(Asking because I met a lovely woman at a party who does this sort of writing, particularly on grant work, as a freelance writer.)
posted by maryr at 9:17 AM on December 19, 2014 [2 favorites]

This is my surprised face.

:|
posted by 1adam12 at 9:21 AM on December 19, 2014 [1 favorite]

I would imagine there is a significant gulf between "bought a paper and made minor adjustments" and "hired a technical writer to produce final documentation for our research," but I'm not involved in this field.
posted by filthy light thief at 9:24 AM on December 19, 2014

Tbh, its not limited to China. I've seen eccentric turns of phrase show up in (random small language group country academia) articles in English - though I suspect that's more to do with "Welp, nobody in our little region reads random blogs in English anyhoo" than this industrial scale gaming.

The 'market' forces are the same however, too much emphasis on publication in the right journals creating incentives for the ambitious and morally ambiguous.

Thank god, I keep my eccentric turns of phrase, the irony is that they might be Hobson Jobson.
posted by hugbucket at 9:27 AM on December 19, 2014

er, that wasn't meant to happen.
posted by hugbucket at 9:27 AM on December 19, 2014

What is the view of having your research written up by an outside source?

If you did the research, got help on the writing, and it's properly acknowledged, it's not considered to be a problem.

In this case, it's not acknowledged -- and the research often isn't done by the authors. They pay for the privilege of getting listed as an author.

In some cases, it's pretty clear that the research isn't getting done at all, or is of such low quality that it can be considered fraudulent.

Ghostwriting/guest authorship is a problem in the US, too, but the dynamics are somewhat different. In the worst cases, money flows to the guest authors rather than from them. (e.g. prominent example from a few years ago.).
posted by cgs06 at 9:28 AM on December 19, 2014 [2 favorites]

Question I don't know the answer to: is there a TurnItIn type process (autocheck for plagiarism) in use at any journals? Can there be, with so much text behind paywalls? Seems like something one of the digital science startups might've targeted, but I can't remember ever seeing anything like that.

I hate saying it, but "assume good intent" on the part of the submitter is pretty much a relic of a different age. When significant parts of the US university system keep downplaying plagiarism, paper buying, etc. at an earlier stage in people's education/careers, especially from foreign students, it shouldn't be a surprise to the academy that this is what we reap.
posted by deludingmyself at 9:31 AM on December 19, 2014

When significant parts of the US university system keep downplaying plagiarism, paper buying, etc. at an earlier stage in people's education/careers, especially from foreign students,

I'm curious about this. At my university--in the departments I've taught for--plagiarism is dealt with harshly, regardless of who commits it. Incoming students are also required to attend a plagiarism/academic honesty workshop so ignorance can't be used as an excuse.

But I've heard this before and I'd like to read more. It seems to be a real concern, and I might have just lucked out.
posted by Kutsuwamushi at 9:58 AM on December 19, 2014

TurnItIn can find when my students plagiarize from paywalled journal articles, and it is mentioned in the article that the publishers are using something similar.
posted by hydropsyche at 10:03 AM on December 19, 2014

TurnItIn can find when my students plagiarize from paywalled journal articles, and it is mentioned in the article that the publishers are using something similar.

Did I miss that? I didn't see anything about routine analysis, just the post facto investigation.
posted by deludingmyself at 10:07 AM on December 19, 2014

Kutsuwamushi, if you're interested in the (tangential to the FPP) topic of universities casting a blind eye to student plagiarism by international students specifically you might want to read through the question and many answers in this recent AskMe. Obviously that's a bunch of strung together anecdotes, but there's a pretty good discussion that got going on the dynamics of how and why this is happening. (The short and cynical version is that foreign students pay full tuition.)
posted by deludingmyself at 10:11 AM on December 19, 2014 [4 favorites]

From the FPP:
BMC, Public Library of Science and other the publishers use plagiarism-checking software to try to cut down on fraud. Software, however, doesn't always solve the issue of plagiarism in journals, Patel warns, paper mills “add another layer of complexity to the problem. It's very worrying."
posted by hydropsyche at 10:11 AM on December 19, 2014

Thanks hydropsyche. I read this first thing this morning and must've missed it. I'm surprised the example given (the "lead to our better, comprehensive understanding" block of text) didn't get flagged, but I suppose the implication is that midsized & smaller journals like Diagnostic Pathology might not be using it.
posted by deludingmyself at 10:15 AM on December 19, 2014

The numbers mentioned in the SA article are small compared to the number of publications per year in the peer-reviewed scientific literature, which is close to 2 million a year, so "widespread fraud" may not be accurate.

However, the phenomenon is perfectly plausible. There are, I am ashamed to say, many incompetents who want to be scientists and professors and have no shame in using any methods at their disposal to gain credibility and promotions. Peer-review is a method to screen publications for suitability. It has a sensitivity and specificity, and no screening method is perfectly specific. A few hundred bad papers out of the many millions submitted each year is a pretty good specificity.

Regarding the fake peer-review, I have been an associate editor for a few journals (one with an impact factor of 15.161), and the reviewers I selected were all known to me by their work. This is the way most reputable journals work. I'm skeptical that the fake reviewer problem is large or affects highly respected journals. It requires a great deal of gullibility on the part of the editors and a willingness to accept reviewers who are unknown in the field.
posted by Mental Wimp at 10:28 AM on December 19, 2014

If a paper has all Chinese authors and is in a journal you've never heard of, take anything it says with a grain of salt. I was taught this early on in my research career, in an environment with scientists from a dozen different countries, including China.
posted by dephlogisticated at 11:18 AM on December 19, 2014 [1 favorite]

maryr: Just a clarification from reading the Scientific American piece - what is the view of having your research written up by an outside source? You legitmately do the research, you make the conclusions, but your communications skills or grammar are lousy, so you hire a third party to write the paper. As long as you acknowledge the third party as an author, even if they did none of the bench or whiteboard, is that OK?

I think it depends on the degree of assistance with the writing they provide.

If you write it up badly and they just write it up better, but the research is all yours, that's fine.

If you pass them an outline with all of the points made and all of the research sources specified, that's probably all right.

If they're doing any of their own subject research and adding any sources, they're an author and have to be included in the authorship.

If you did the lab work and the analysis but they did all of the research and writing (i.e. you passed them the experiment but not much else), they're a major author. Second for sure, maybe first (you should take final author if you are the PI).

If they did any of the analysis plus all of the research and writing (i.e. you just passed them the data and maybe a partial analysis), they're the first author for sure.
posted by Mitrovarr at 11:25 AM on December 19, 2014

You legitimately do the research, you make the conclusions, but your communications skills or grammar are lousy, so you hire a third party to write the paper.

Why would anyone care who reported the findings as long as the research was genuine and was genuinely yours?
posted by pracowity at 12:54 PM on December 19, 2014

Well, someone from an institution with less money might be upset that you could afford a writer, I suppose. I'm not sure if you'd be allowed to spend grant money that way, for example. But it does seem like a reasonable way of doing business - I don't think anyone would complain if you hired someone to do an anatomical drawing for you.

Less a validity complaint or polluting the literature and more a "those damn folks at Harvard" kind of complaint. (My sources suggest that Harvard gives shitty startup packages to junior faculty, so those damn Harvard folks may not be able to afford writing help either.)
posted by maryr at 1:12 PM on December 19, 2014

I hate saying it, but "assume good intent" on the part of the submitter is pretty much a relic of a different age

That is an excuse translates to roughly, "We have the power to do what we want and we don't want to question ourselves."

And people wonder why systems become dysfunctional...
posted by Alexandra Kitty at 2:36 PM on December 19, 2014

pracowity: Why would anyone care who reported the findings as long as the research was genuine and was genuinely yours?

Because there's more to science than just physically doing experiments. It takes a lot of book-research and integrative analysis to write a good introduction and discussion, and a lot of familiarity with the experiment to write the materials and methods and the results. I can't imagine how you could have someone else write it all for you without either having them so do much research that they were legitimately part of the authorship, or handing them something so completely written it was basically a first draft and they're just doing editorial work.
posted by Mitrovarr at 2:38 PM on December 19, 2014 [4 favorites]

I remain amazed that during the manuscript submission process, I am still, in this day and age of captchas, asked for the email addresses of the reviewers I suggest. This is so ripe for misuse. It may work for upper-echelon journals, and even the trade journals, wherein editors know the players and can easily recognize names (and presumably spot and deal with fraud) at a glance, but it was obviously going to be exploited.

Just this week, I was requested by an upstart, but well recognized (IF 4 or so) open-access journal in my field to review a ms from foreign authors (not Chinese, but not the usual set of countries). It looks like the submission has been hot-potato'd around since August. It should be obvious that, aside from being in *ology in general, I do not work in their field, so I do wonder why I was asked. It could be that the authors are just so unrecognized AND illiterate within the field that no one, including they or the intern editor, could come up with suitable reviewers' names (I did not see the ms, so cannot comment on their bibliography). But it also could be, well, a hot potato.
posted by Dashy at 3:35 PM on December 19, 2014 [1 favorite]

Happens in the humanities too. I've been approached by a Chinese colleague who wanted access to a prestigious US journal. He was offering me a spot as an "editor" of a special issue containing contributions by his friends; apart from lending my name I would have had to do no work at all. I said no and he moved on to another colleague.
posted by homerica at 2:58 AM on December 20, 2014

Related: Professors Maggie Simpson and Edna Krabappel
posted by Mental Wimp at 11:44 AM on December 22, 2014

Here's a doozy.

"We retracted the paper because of an error in one of (the) data points on the BMI graph and because, as the FTC pointed out to us, there was inadequate disclosure of diet restrictions on the subjects and inadequate disclosure of the blinding procedures for the supplements given the subjects," Vinson and Burnham said in a joint statement Thursday.

When scientists who author peer-reviewed publications need to be instructed by the FTC(!) on proper conduct of research, I have no words. None.
posted by Mental Wimp at 10:38 AM on January 5, 2015

It's a weird one, but the FTC and FDA have shared jurisdiction on regulating dietary supplements. A piece of 1994 legislation mandated that FDA treat dietary supplements as foods, not drugs, so in some ways the FTC actually has more authority than the FDA on this one.
posted by deludingmyself at 11:29 AM on January 5, 2015

Yep, but my point was that the FTC is not a bastion of scientific integrity; they need to set the bar pretty low to avoid being overwhelmed with regulatory actions. The two authors are supposedly rigorous scientists (one in chemistry and the other in psychology at Scranton University) whom the FTC should be consulting, not the other way around. In a way, this reveals them to be, if not frauds, at least incompetent.
posted by Mental Wimp at 3:02 PM on January 5, 2015