A Cornell professor whose buzzy and accessible food studies made him a media darling has submitted his resignation, the school said Thursday, a dramatic fall for a scholar whose work increasingly came under question in recent years. The university said in a statement that a year-long review found that Brian Wansink "committed academic misconduct in his research and scholarship, including misreporting of research data, problematic statistical techniques, failure to properly document and preserve research results, and inappropriate authorship."
Wansink, a marketing professor at Cornell's business college who was the director of the university's Food and Brand Lab, will retire at the end of the academic year, the school said. The move follows the recent retraction of six of Wansink's papers by the American Medical Association's JAMA Network, including those about how serving bowl size affected food consumption, how fasting changed people's food preferences and how action-packed television programs increased food intake.
Wansink emailed The Washington Post on Thursday a news release of his retirement, which included statements attributed to a university trustee saying that "Cornell and Professor Wansink mutually have decided that Professor Wansink's research approach and goals differ from the academic expectations of Cornell University, and they have decided to part ways accordingly." Wansink said he is leaving his position June 30, 2019.
For years, Wansink enjoyed a level of prominence that many academics would strive for, his work spawning countless news stories. He published a study showing that people who ate from "bottomless" bowls of soup continue to eat as their bowls are refilled, as a parable about the potential health effects of large portion sizes. Another, with the title "Bad popcorn in big buckets," similarly warned about the perils of presenting food in big quantities, according to Vox.
Also at Science Magazine, Ars Technica, CNN, and Retraction Watch.
(Score: 3, Informative) by pTamok on Monday September 24 2018, @05:53PM (8 children)
If you read this chap's blog articles, but start first with the paper he published with two others:
Statistical heartburn: an attempt to digest four pizza publications from the Cornell Food and Brand Lab [biomedcentral.com]
Nick Brown's Blog: A different set of problems in an article from the Cornell Food and Brand Lab [blogspot.com]
Cornell salutes America's teenage female combat heroes of WW2 [blogspot.com]
Nick Brown's Blog: More apparent duplication from the Food and Brand Lab [blogspot.com]
Nick Brown's Blog: Strange patterns in some results from the Food and Brand Lab [blogspot.com]
Nick Brown's Blog: More problematic articles from the Food and Brand Lab [blogspot.com]
Nick Brown's Blog: The final (maybe?) two articles from the Food and Brand Lab [blogspot.com]
Nick Brown's Blog: Problems in Cornell Food and Brand Lab's replacement "Can Branding Improve School Lunches?" article [blogspot.com]
Nick Brown's Blog: The latest Cornell Food and Brand Lab correction: Some inconsistencies and strange data patterns [blogspot.com]
Nick Brown's Blog: The Cornell Food and Brand Lab story goes full circle, possibly scooping up much of social science research on the way, and keeps turning [blogspot.com]
It took a long time, and a lot of patient work to challenge Wansink's published articles.
(Score: 3, Interesting) by takyon on Monday September 24 2018, @06:20PM (7 children)
Publish everything somewhere (arXiv, university sites, etc.) and let other scientists "peer review" it (publicly or anonymously, but with all such comments printed somewhere).
If a journal is taking money to publish papers, and reviewers aren't sufficiently motivated, you have a recipe for false peer review.
Scientists who disagree with a paper's conclusions will attack it in order to disprove it. Scientists who agree can try to replicate the results.
[SIG] 10/28/2017: Soylent Upgrade v14 [soylentnews.org]
(Score: 4, Insightful) by quietus on Tuesday September 25 2018, @07:04AM (5 children)
Thoroughly checking a paper takes time, and effort.
You're gonna end up with the open-source conundrum [securityledger.com]: lots and lots of people assuming someone else already has done the proper vetting, especially when there's a reputation involved.
(Score: 2) by takyon on Wednesday September 26 2018, @12:30AM (4 children)
If most papers are ineffectively reviewed bullshit about obscure topics, then they don't matter much anyway. A certain percentage of content could be ignored. Maybe over 90%?
Controversial topics, like Planet Nine, EmDrive, dark matter, etc. will have no shortage of scientists looking to disprove what others have put forward.
Government agencies like the FDA and NIH will mandate the replication of medical studies. Replication of the results can act as a review (previous studies can be critiqued within the text of a new study). Governments can mandate peer review [wikipedia.org] for science that is used for policymaking. Apparently, the State of California already does this.
If academic scholars are already peer reviewing papers as part of their job, they can continue to do so even under a new system. They weren't paid directly by the journals anyway.
The scientific method predates the scientific journal. Just share your ideas, and they will either stand the test of time and scrutiny or not. The peer review can still happen, just not as part of the opaque and corrupt science journal process. Comments can be published anonymously. That might not defeat style analysis, but it's on the reviewer to manage that risk if they want to participate in a transparent peer review process.
Even if you do nothing to change how journals and peer review work, the lack of replication of studies is already a recognized problem:
https://en.wikipedia.org/wiki/Replication_crisis [wikipedia.org]
https://www.bbc.com/news/science-environment-39054778 [bbc.com]
https://blog.frontiersin.org/2017/05/01/are-replication-studies-unwelcome/ [frontiersin.org]
[SIG] 10/28/2017: Soylent Upgrade v14 [soylentnews.org]
(Score: 0) by Anonymous Coward on Wednesday September 26 2018, @09:00PM
Is this a joke?
(Score: 2) by quietus on Thursday September 27 2018, @11:35AM (2 children)
You will still need some kind of authority to distinguish between the quality, or insights* if you like, of different papers; even if something comparable to soylentnews.
* Acknowledging ofcourse, that new insights tend to be disputed at first, i.e. if you do not define strict quality criteria, you'll likely end up with as many upmods as downmods.
(Score: 2) by takyon on Thursday September 27 2018, @01:39PM (1 child)
Wind the clock back to the early days of science. Most sharing of results was in letters to other scientists, or presentations. There wasn't any formalized peer review, authoritative gatekeeper, or electronic voting.
Today, it would be simple to just produce a Stack Overflow clone for discussion and voting. Or you could design a platform from scratch.
[SIG] 10/28/2017: Soylent Upgrade v14 [soylentnews.org]
(Score: 2) by quietus on Sunday September 30 2018, @10:06AM
It might be better to start from a platform akin to PLOS One and/or Scientific Reports. Just looking at PLOS One, I think there might be three areas of improvement.
A further improvement could consist of adding tools, dedicated to making reviews easier -- automated statistical checks, extracting and (dis)aggregating experimental data, and so on.
(Score: 1, Insightful) by Anonymous Coward on Wednesday September 26 2018, @02:51PM
Yes, in the limit of infinite time and resources, but in practice, no. Claim to break the laws of physics? Yes, there is a lot of motivation there. Claim that bigger bowls cause you to eat more? Well, I don't know if I believe that, but I'm not dropping what I'm doing to try and replicate that (unless some funding source wants to give me money to do that).