Maybe you’re a researcher who likes keeping up with developments in scientific integrity. Maybe you’re a reporter who has found a story idea on the blog. Maybe you’re an ethics instructor who uses the site to find case studies. Or a publisher who uses our blog to screen authors who submit manuscripts — we know at least two who do.
An endocrinology journal has pulled a 2017 paper by a group from Russia and Romania because, well, maybe it’s just better if you read for yourself.
The article, “Testosterone promotes anxiolytic-like behavior in gonadectomized male rats via blockade of the 5-HT1A receptors,” appeared in General and Comparative Endocrinology, an Elsevier publication.
How common are calculation errors in the scientific literature? And can they be caught by an algorithm? James Heathers and Nick Brown came up with two methods — GRIM and SPRITE — to find such mistakes. And a 2017 study of which we just became aware offers another approach.
Jonathan Wren and Constantin Georgescu of the Oklahoma Medical Research Foundation used an algorithmic approach to mine abstracts on MEDLINE for statistical ratios (e.g., hazard or odds ratios), as well as their associated confidence intervals and p-values. They analyzed whether these calculations were compatible with each other. (Wren’s PhD advisor, Skip Garner, is also known for creating such algorithms, to spot duplications.)
After analyzing almost half a million such figures, the authors found that up to 7.5% were discrepant and likely represented calculation errors. When they examined p-values, they found that 1.44% of the total would have altered the study’s conclusion (i.e., changed significance) if they had been performed correctly.
We asked Wren — who says he thinks automatic scientific error-checkers will one day be as common as automatic spell-checkers are now — to answer a few questions about his paper’s approach. This Q&A has been slightly edited for clarity.
After years of back and forth, a highly cited paper that appeared to show that gay people who live in areas where people were highly prejudiced against them had a significantly shorter life expectancy has been retracted.
Retraction Watch readers may recall the name Yoshihiro Sato. The late researcher’s retraction total — now at 51 — gives him the number four spot on our leaderboard. He’s there because of the work of four researchers, Andrew Grey, Mark Bolland, and Greg Gamble, all of the University of Auckland, and Alison Avenell, of the University of Aberdeen, who have spent years analyzing Sato’s papers and found a staggering number of issues.
Those issues included fabricated data, falsified data, plagiarism, and implausible productivity, among others.In 2017, Grey and colleagues contacted four institutions where Sato or his co-authors had worked, and all four started investigations. In a new paper in Research Integrity and Peer Review, they describe some of what happened next:
When it comes to plagiarism, there is apparently no statute of limitations.
That’s one lesson one might take from this tale of two papers, one published in 1984 in the American Journal of Obstetrics and Gynecology (AJOG), and the other published in 2000 in the Medical Journal of The Islamic Republic of Iran (MJIRI). Both are titled “The use of breast stimulation to prevent postdate pregnancy.”
Imagine you’re a journal editor. A group of authors sends you a request to retract one of their papers, saying that “during figure assembly certain images were inappropriately processed.”
What do you do next? Do you ask some tough questions about just what “inappropriately processed” means? Do you check your files for whether the author’s institution had told you about an investigation into the work? Do you Google the author’s names? Do you…search Retraction Watch?
It seems unlikely that any of those things happened in the case of a recent retraction from Nature Communications, or, if they did, they don’t seem to have informed the notice. We don’t know for sure, because, as is typical, the journal isn’t saying much. But here’s what we do know.Continue reading Why journal editors should dig deeper when authors ask for a retraction
A journal has retracted a paper on a drug for a blood disorder 20 years after it was published — and 17 years after an author of the article was told to request the move by his university.