Fresh concerns raised over academic conduct of major US nutrition and behaviour lab

With eleven publications previously under scrutiny, new manifestation develops of duplicate pamphlet and data oddities in the work of Professor Brian Wansink

The head of Cornell Universitys Food and Brand lab is facing reincarnated allegations regarding academic impropriety, including self-plagiarism and possible data misrepresentation. Professor Brian Wansink, who has authored hundreds of scientific papers and is a former agency director in the US Department of Agriculture, is famous for promoting the concepts of mindless eating and the idea that people find it easier to control their food intake when dining from smaller slabs. However, investigations conducted by University of Groningen PhD student Nick Brown have clearly revealed echoed cases of duplicate pamphlet in Wansinks research, as well as odd data oddities across two analyzes.

Wansinks research first is under the responsibility of scrutiny in late 2016 when, in a blog announce called The Grad Student Who Never Said No, he appeared to champion the use of gray-headed experiment practices as vocation tools for young scientists practices such as selectively reporting positive results from a dataset of primarily null sequels, and presenting data fishing as hypothesis testing. Analyses of the published answers by Tim van der Zee, Jordan Anaya and Nick Brown afterwards identified what appear to be hundreds of statistical incompatibilities in four members of the articles where Wansink acknowledged deploying these practices, and Anaya afterwards raised concerns about the accuracy of seven additional publications. After was unwilling to share the raw data, Wansink promised to conduct an internal investigation of four of the 11 publications.

In a new investigation published today on his personal blog, Brown now reports numerou cases of apparent duplicate pamphlet: instances where massive amounts of textbook in Wansinks sections recur, practically verbatim, the process of drafting other sections he authored, without announced today that recycling took place, and despite the articles being scheduled as independent publications. Undeclared text recycling, also known as self-plagiarism, can be considered academic impropriety when played on a large scale, and where reference is includes as in these cases text other than the descriptions of research methods.

Brown says he uncovered the cases of duplication while attending a separate investigation. When I was writing a previous blog announce on a research essay from this laboratory, I went go looking for awards of that essay, in case a correction had been issued, he announced. I noticed that Dr. Wansink had cited the essay twice, exploiting almost identical utterances. As I examined the two quoting sections more closely, I noticed that quite a number of the textbook was reproduced between them.

As well as reporting several instances of where textbook was reproduced, Brown too alleges a even more serious dispute that, at a minimum, hints an extraordinary co-occurrence, but may point to repetition pamphlet of data, or might also indicate data manipulation. Brown claims to show that two analyzes published by Wansink in 2001 and 2003 present uncannily similar answers, with 39 out of 45 sequels identical to the decimal point, despite being drawn from different samples. In the 2001 investigate, Wansink reports recruiting 153 representatives from the Brand Revitalisation Consumer Panel while in the second study the reported sample is constituted by 654 respondents to a national overlook based on addresses obtained from census enters. How such similar answers could emerge from two definite analyzes, and two definite samples, continues unexplained. At the time of pamphlet, Wansink had not responded to requests for statement. A inquiry to the Cornell University Office of Research Integrity and Assurance resulted in a reply from Finn Partners PR Firm, pointing to a statement from Wansink that did not address the most recent allegations.

Unusual
Unusual similarities uncovered by Brown between the reported results of two papers, published by Brian Wansink in 2001( left) and 2003( right ). The highlighted numbers are identical between the two tables, although there are the studies are reportedly conducted on different samples of people. Composite: Nick Brown

Brown believes the problems he has disclosed warrant further investigation. What happens next with regard to these publications is up to the publishers and journal journalists, he announced. He also argues that Cornells Office of Research Integrity and Assurance should encourage the university to liberate the data sets for the various previous analyzes where he, van der Zee and Anaya marked countless statistical incompatibilities, including some answers that are mathematically hopeless. At the same time, Brown is unmoved by suggests that he has targeted Wansink unfairly. All Ive tried to do here is to bring to light what turns out to be points, supported by evidence that can be quite easily are audited by anyone who maintenances to gape, he announced. I think thats part of lawful dialogue in science.

Below is the statement received without Finn Partners PR Firm on 28 February 2017, which was published by Brian Wansink on Cornell Universitys website on 13 February 2017 :

Recent wonders have arisen seeing statistical approaches which I implement in my experiment. I, along with my colleagues at Cornell Universitys Food and Brand Lab, make these questions very seriously. We are currently attending a full review of studies in question, organizing comprehensive data which will be shared and establishing new standards for future operations at the lab which will include how we respond to requests for experiment report. Over the next few months, I will focus on the efforts are indicated below and pass full attention to this important matter. Once this review is end, I will share my findings with my peers and the media. In the meantime, we will continue to conduct the research and outreach design that is the signature of Cornells Food and Brand Lab .

Read more: https :// www.theguardian.com/ discipline/ head-quarters/ 2017/ tainted/ 02/ fresh-concerns-raised-over-academic-conduct-of-major-us-nutrition-and-behaviour-lab