Category: data issues

Mega-correction for “empirical anomalies” in management paper

The Academy of Management JournalThe author of a paper that looked at how the geographical spread of research and development sites has impacted innovation has posted a four-page list of corrections that fixed “empirical anomalies” in the paper.

A group of PhD students raised concerns about the paper’s findings, according to the editor-in-chief of The Academy of Management Journal, Gerard George. The journal formed a committee that worked with the author to reproduce the results. That ended with a correction to two of the paper’s three hypotheses, and corresponding parts of the text.

The four-page notice — (the details of which are paywalled, unfortunately) — includes notes from the journal’s editor and the author:

Continue reading Mega-correction for “empirical anomalies” in management paper

Activist group retracts warnings about midwest oil wells

downloadAfter receiving additional data from the government, an activist group has retracted an analysis that suggested energy companies were not taking steps to cut back on a controversial practice.

The Bakken analysis — named for North Dakota’s gigantic underground deposit of oil and natural gas — was published by the Environmental Law and Policy Center (ELPC). It focused on a practice known as “gas flaring” — burning natural gas instead of using or selling it. The analysis, released last month, found that hundreds of wells in North Dakota had not filed the necessary plans for saving excess gas produced in the course of extracting oil from wells. But after the Department of Mineral Resources provided more data — we’re not sure what kind of data, specifically — the ELPC retracted that conclusion.

The group posted the retraction notice on September 25, just four days after they presented the analysis at a news conference. It states why their recent evaluation of companies’ plans to capture excess gas might be wrong — and what they’re doing next:

Continue reading Activist group retracts warnings about midwest oil wells

Boldt’s retraction count upped to 94, co-author takes legal action to prevent 95th

Screen Shot 2015-09-16 at 12.12.13 AM

We’ve found two recent retractions and an expression of concern for Joachim Boldt, former prominent anesthesiologist and currently Retraction Watch leaderboard’s 2nd place titleholder. He now has 94 retractions.

One of the retracted articles contains falsified data, along with a researcher who didn’t agree to be a co-author, according to an investigation by the Justus Liebig University Giessen, where Boldt used to work. The expression of concern is regarding some questionable data. The other new retraction is actually one of 88 papers that a group of editors agreed to retract back in 2011, after they were “unable to verify” approval by the Institutional Review Board (IRB) for the studies.

One of those 88 papers, we’ve discovered, has still has not been retracted. According to an editor at the journal, they haven’t removed it because one of Boldt’s co-authors has threatened them with legal action. Continue reading Boldt’s retraction count upped to 94, co-author takes legal action to prevent 95th

13th retraction issued for Jesús Ángel Lemus

Proceedings of the Royal Society B- Biological SciencesA 13th retraction has been published for Jesús Ángel Lemus, the Spanish veterinary researcher whose work colleagues have had trouble verifying.

This paper was pulled for similar reasons as his other retractions: After retrying the experiments in two independent labs, fellow authors were “unable to arrive to any sound conclusion about the validity of his analyses.” 

Proceedings of the Royal Society B: Biological Sciences posted the notice September 16, three years after the paper received an expression of concern.

The retraction notice, signed by every co-author but Lemus, reads:

Continue reading 13th retraction issued for Jesús Ángel Lemus

Authors “did not have permission” to use pesticide data

10661An environmental journal is retracting an article about the risks of pesticides to groundwater after determining it contained data that “the authors did not have permission (implicit or explicit) to publish.”

According to the retraction note in Environmental Monitoring and Assessment, the paper said the data came from a non-author’s PhD thesis, but it’s not there. Those mysterious data were used to validate a model for pesticide exposures, described in an excerpt from the abstract:  Continue reading Authors “did not have permission” to use pesticide data

Cochrane withdraws review on zinc for colds for data concerns

Cochrane_LogoThe Cochrane Library has withdrawn a 2013 systematic review on zinc’s ability to fight the common cold.

Cochrane often marks reviews “withdrawn” once new evidence emerges that renders them out of date — but in this case, the review was flagged while the editors investigate issues “regarding the calculation and analysis of data.”

Here’s the notice.

Continue reading Cochrane withdraws review on zinc for colds for data concerns

Investigation ends in 6th retraction for Voinnet

PLOS PathogensA sixth paper co-authored by plant researcher Olivier Voinnet has been retracted by PLOS Pathogens “following an investigation into concerns.”

The investigation found “several band duplications” in one figure provided by fifth author, Patrice Dunoyer, who took it from “the Master thesis of a former student working under his supervision, without the prior consultation or consent of this student,” according to the notice. There was also an incorrect “loading control” in another figure, attributed to first author Raphael Sansregret and last author Kamal Bouarab. 

Voinnet and Bouarab, the study’s corresponding authors, took full responsibility for “the publication of this erroneous paper.”

Although investigators found that the raw data in the duplicated figure backed up its conclusions, “given the nature and extent of data manipulation,” the authors asked the journal to retract the paper .

Continue reading Investigation ends in 6th retraction for Voinnet

Investigation finds “careless data workup” in alcoholism drug paper

S00142999

An investigation at Karolinska Institute has led to the retraction of a paper about drug treatments for alcoholics, after concluding the article contains a “very careless data workup.”

The paper, “Memantine enhances the inhibitory effects of naltrexone on ethanol consumption,” found that the drug memantine (normally used to treat Alzheimer’s) enhances the effects of naltrexone in rats, which blocks the high of alcohol.  It was published in the European Journal of Pharmacology and has been cited 10 times, according to Thomson Scientific’s Web of Knowledge.

However, its conclusion is now “unreliable,” according to the retraction note:

Continue reading Investigation finds “careless data workup” in alcoholism drug paper

Marine mammal injury study retracted for using restricted gov’t data

Integrative ZoologyA 2012 paper that analyzed injuries to aquatic mammals in China has been retracted “due to the usage of restricted data from the Ministry of Agriculture of China.”

The authors — from Shandong University in China, The University of Hong Kong and the Peruvian Centre for Cetacean Research — “organized the collection of official documents related to strandings, bycatches and injuries of aquatic mammals in the waters of mainland China from provincial fishery administrations for the years 2000 to 2006,” according to the abstract. However, they may not have been supposed to do that.

Here’s the retraction notice from Integrative Zoology (which is paywalled, tsk, tsk):

Continue reading Marine mammal injury study retracted for using restricted gov’t data

Journal corrects CrossFit injury data in paper at center of lawsuit

XLargeThumb.00124278-201509000-00000.CVA study on the trendy and grueling workout regimen known as CrossFit has a correction concerning the number of participants hurt during 10 weeks of training. The paper has been the center of multiple lawsuits  — one by CrossFit, and one by a CrossFit gym owner — for allegedly over-inflating the risks associated with the regimen.

The original paper claimed that 9 of 54 participants dropped out of the study due to “overuse or injury.” The correction note says that just two left for those reasons.

The paper, published in The Journal of Strength and Conditioning Research actually concludes that CrossFit has some benefits. According to the abstract: Continue reading Journal corrects CrossFit injury data in paper at center of lawsuit