The Mysterious Case of the Two Papers With Identical Results

Get your science fix here: research, quackery, activism and all the rest
Post Reply
User avatar
Fishnut
Snowbonk
Posts: 548
Joined: Mon Nov 11, 2019 1:15 pm

The Mysterious Case of the Two Papers With Identical Results

Post by Fishnut » Fri May 29, 2020 2:46 pm

@MicrobiomDigest has a very interesting thread looking at two papers researching different cancers which have identical results. The papers are from different sets of authors, at different institutes, published around the same time. The thread is well worth a read, not least for the 'twist' at the end.

A reply highlights an article written in Science in 2014 about copycat papers being produced on demand by companies charging thousands of dollars a time.

To try and start a discussion I'll ask: what can be done to prevent this? When there's so many papers being produced it's impossible to keep a systematic eye on them and spot duplication. AI doesn't seem to be up to the task. Are there social forces we can use? I know there are cultural differences in understanding of what constitutes plagiarism but this isn't cultural differences, this is outright fraud.

User avatar
Bird on a Fire
Stummy Beige
Posts: 2950
Joined: Fri Oct 11, 2019 5:05 pm
Location: nadir of brie

Re: The Mysterious Case of the Two Papers With Identical Results

Post by Bird on a Fire » Fri May 29, 2020 3:02 pm

So over 400 papers from the same paper mill, all identified because they share the same text and figures.

https://twitter.com/MicrobiomDigest/sta ... 69697?s=20

Fraud in published science is obviously even bigger than we thought. Student essays have to go through Turnitin but obviously whatever post-publication checks are done on actual publications are hopelessly inadequate.

It certainly means that you can't trust something just because it's been published. If such unsophisticated fraud is so widespread subtler malpractice must be rife.

Clearly we need automated tools to compare everything against everything else before final versions are posted online.

And I can't imagine the big commercial publishers are too bothered, which is yet another reason for-profit publication needs to die. My group only publishes in society journals, which is ethically good, but does rule out a lot of the higher-impact ones - scientists shouldn't have to choose between subsidising abusive practices and career progression.
Born at 356.32 ppm CO2

User avatar
Fishnut
Snowbonk
Posts: 548
Joined: Mon Nov 11, 2019 1:15 pm

Re: The Mysterious Case of the Two Papers With Identical Results

Post by Fishnut » Fri May 29, 2020 3:22 pm

Bird on a Fire wrote:
Fri May 29, 2020 3:02 pm
And I can't imagine the big commercial publishers are too bothered, which is yet another reason for-profit publication needs to die.
This is, I think, key to the issue. The big publishers don't care as they get money regardless, the peer-reviewers don't have any incentive other than honour to do a good job and have few resources to thoroughly check that results are reported accurately, and there's such a bombardment of new papers that it's easy for most people to miss that a bunch of papers look identical.

I do think it speaks to the broader issues with the publish-or-perish mentality that much of science has fallen into, and I'm not sure how we fix that.

Post Reply