As the embattled director of Cornell University’s Food and Brand Lab continues to face heavy criticism from researchers who accuse him of publishing papers with scores of data inconsistencies, students who worked in the lab told The Sun on Wednesday that the director’s retracted work stains the University’s reputation — and their own.
Prof. Brian Wansink, marketing, retracted a fifth paper after editors said the 2003 paper cited “unreliable data” and text duplication from an earlier soy consumption study co-authored by Wansink.
The retraction of the study by the journal Appetite marked the sixth retraction for Wansink — he retracted one study twice. Wansink, who founded the Food and Brand Lab in 1997 at the University of Illinois and moved it to Cornell in 2005, has also issued or is planning to issue eight total corrections, BuzzFeed reported.
When asked for comment, Cornell pointed to an October statement from Joel Malina, vice president for University relations, in which he said Cornell is taking “the questions raised about Professor Wansink’s work quite seriously.”
Cornell has launched an internal investigation “in compliance with our internal policies and any external regulations that may apply,” Malina said in the October statement.
John Carberry, the senior director of media relations, declined to comment on when the investigation began, what other studies are being examined or whether the investigation involves the Office of Research Integrity.
In April, Cornell conducted an internal investigation of Wansink’s publications and determined in initial reviews that while Wansink handled data inappropriately, his errors “did not constitute scientific misconduct,” The Sun previously reported. It’s unclear why that investigation failed to uncover the scope of Wansink’s questionable research.
Ellen Ransley ’19 said in an interview that she left the lab after a year of working as a research assistant in the 2017-2018 academic year as researchers began to question the lab’s research.
“I didn’t want to be associated with it,” she told The Sun.
Bill Wingell / The New York Times
Using a one-way window at the Cornell Food and Brand Laboratory, director Brian Wansink keeps an eye on students taking part in an eating behavior study in October 2006.
Wansink “is very well established,” Ransley said. “The people that worked under him read the papers he published and didn’t question the data and conclusions. They assumed it was correct because he’s been working in field for a long time.”
Ransley said Wansink often encouraged members of the lab to produce as many observations and variables as possible for a given study, which she said often led to researchers chasing down unreliable conclusions.
“More people need to review his papers and spend more time actually reviewing his data,” she said, noting that some people believe Wansink, who has tenure, will have to leave Cornell. “He should hire people to specifically go over data … a lot of what he did was very independent.”
A Cornell alumna who worked in the Food and Brand lab as an undergraduate in 2013 said members of the lab drew broad conclusions based solely on correlations. The alumna, who spoke on the condition of anonymity because she feared professional repercussions, said lab members portrayed these correlations as causations to journal editors and the news media.
“We already had the result that we wanted in our minds,” the alumna told The Sun, “and many of the studies the fellow students and I were involved in were observational.”
Studies testing whether overweight people sit closer to buffet tables and whether people tend to wear stretchier clothing at fast food restaurants led researchers to seemingly obvious — but unreliable — conclusions, she said.
“The conclusions we were aiming to draw were ‘sit far from the buffet’ and ‘wear jeans,’ but this is quite the jump,” she said in an email.
The alumna said that whenever Wansink’s name comes up, “I sigh and think, ‘but what did the numbers actually say?’”
Courtesy of Jason Koski / Cornell University
Prof. Brian Wansink, director of the Cornell Food and Brand Lab.
Wansink, who holds degrees in business administration, journalism and marketing, according to his CV, has also been criticized for not holding a degree in food science.
Dean Hauser ’17, a food science major, told The Sun he feels frustrated as a Cornell food science alumnus “pursuing an advanced degree and having people ask about the Cornell Food and Brand [Lab] scandal when [Wansink] is not even part of the department.”
“Calling him a ‘Cornell Food Scientist’ inaccurately groups his dodgy ‘food psychology’ research with Cornell food science, and hurts the department’s great reputation,” Hauser said in an email.
In March 2017, University of Groningen Ph.D. student Nicholas J. L. Brown noticed that Wansink’s 2003 paper on soy consumption contained text copied from two previous papers Wansink co-authored in 2001 and 2002. Brown said there are “mysterious” statistical patterns in all three papers.
“We have seen what appear to be impossible means and test statistics; inconsistent descriptions of the same research across articles; and recycling of text (and even, it would appear, a table of results) from one article or book chapter to another,” Brown wrote in a blog post.
“The attitudes towards soy products reported by the … studies are remarkably similar, despite the samples having been drawn from very different populations,” he continued. “This similarity also seems to apply to the items for which the results give impossible test statistics.”
Wansink and seven other current members of the lab did not respond to a request for comment from The Sun.
Wansink issued his most recent correction for “Attractive Names Sustain Increased Vegetable Intake in Schools,” a study he published in a 2012 issue of Preventive Medicine.
The original study claimed that its participants were children from 8 to 11 from five elementary schools, but the correction said the children participating in the study were actually preschoolers between 3 and 5, many of whom were enrolled in programs that met in elementary schools.
Wansink and his co-authors responded that “these mistakes and omissions do not change the general conclusion of the paper that attractive names increase vegetable intake in schools across a wide age-range of children.”
Other researchers disagreed.
Brown said the data in the study, “corrected or not, cannot be the basis of any sort of scientific conclusion about whether changing the labels on vegetables makes children want to eat more of them.”
Another researcher and vocal critic who first exposed some of Wansink’s mistakes, Tim van der Zee, said in a tweet that he found it “extremely odd” that Wansink issued a correction instead of retracting the study.
University of Pennsylvania Prof. James C. Coyne, psychology, said on Facebook that Wansink’s initial correction may eventually turn into a full retraction.
“Why does Brian Wansink keep getting ‘corrections’ to his dishonest science, rather than outright retractions?” Coyne said.