Retraction Watch

Tracking retractions as a window into the scientific process

More notices appear for embattled Cornell food researcher

without comments

Journals have posted two corrections alongside papers by Brian Wansink, a food researcher whose work has lately come under fire.

One of the corrected papers was among the initial batch that raised eyebrows last year; after Wansink praised the productivity of one of his researchers, critics suggested four papers contained critical flaws. The questions about his work soon extended to other papers, one of which was retracted in April. That month, an internal review by Cornell University announced that Wansink made numerous mistakes, but did not commit misconduct. Wansink has pledged to reanalyze multiple papers.

One paper that was among those initially criticized has received a formal correction notice from the Journal of Product & Brand Management. The notice, which appears behind a paywall, has drawn fire from a regular critic of Wansink’s work, Jordan Anaya, who argues “this correction actually needs a correction, actually several.”

Here’s the notice for “Peak-end pizza: prices delay evaluations of quality:”

It has come to our attention that the article “Peak-end pizza: prices delay evaluations of quality” David R. Just, Ozge Sigirci and Brian Wansink, published in Journal of Product & Brand Management, Vol. 24 No. 7, pp. 770-778 contains errors in the data presented and does not fully attribute one of the sources drawn upon.

In response to recent criticism of the original work and in addition to institutional reanalysis, the authors have sought the independent feedback of a researcher at Mathematica Policy Research who has in turn reviewed the text, tables, and Stata output contained in this correction for consistency. Mathematica was compensated for this work.

These analyses focused on pizza, therefore diners who did not report eating at least one piece of pizza were not included in the analyses. Consistent with the original manuscript, two additional diners are eliminated from this analysis because one person’s height was noted as 8-inches and another one’s weight was noted as 450-lbs.

Table I originally reported the number of observations for each column erroneously as n 62, n 60 and n 122 for the $4 buffet, $8 buffet and all treatments respectively. In fact, the number of observations varies by question due to respondents skipping questions. The corrected table, below, lists the number of observations for each cell. In addition, there are two other slight changes due to rounding errors. The online version of Table I has been corrected to attribute the original source “Lower Buffet Prices Lead to Less Taste Satisfaction”, published in Journal of Sensory Studies (2014), Vol. 29 No. 5.

In conducting this updated analysis, the authors opted to use Stata 14.0 for convenient scripting and log file generation and to use a Hotelling’s test of differences in ratings rather than the F-test reported in the article. The article reports F-tests for differences in average ratings of first, middle and last slice respectively of 16.56 (p 0.00) and 0.65 (p 0.53) for the $4 and $8 conditions respectively. The Hotelling’s test produces F-statistics of 10.44 (p 0.01) and 0.98 (p 0.39), respectively. Minor differences in rounding were found for Tables II and III. Note that in Table III different numbers of observations are used for each line due to differential response rates to survey questions. A full script and log file can be found here: https://doi.org/10.6077/J5CISER2783.

These errors have been corrected in the online version. The authors apologise sincerely for these errors.

The 2015 paper has been cited once, according to Clarivate Analytics’ Web of Science.

We contacted the journal to ask why the correction had been paywalled. A representative of publisher Emerald Publishing, Richard Whitfield, told us:

The full correction, free from the paywall, will appear in the next Issue of JPBM (In response to Jordan Anaya’s initial concerns, we were keen to provide him with early access to the correction).

We asked if the journal had a response to Anaya’s criticisms (he called it the “worst correction ever”); Whitfield said:

We thank you for your interest but following the latest blog entry by Jordan Anaya, we are unable to comment while we continue our own enquiries.

There’s also been a correction posted for a 2008 paper co-authored by Wansink, which focuses on how people eat at Chinese buffets. Here’s the notice from the journal Obesity:

In the published article above, Table 2 did not specify the number of observations for each variable, as well as missing observations. This table has been revised to clarify that the maximum cell counts for each column are 71, 70, and 72, but that for the depicted variables, they can range from 64-71, 66-70, and 68-72. The authors do not feel the small changes in the table below impact the conclusions of the research.

Eating behavior and obesity at Chinese buffets” has been cited 26 times since it was published in 2008. (Anaya has also shared his thoughts on this correction, and flags a video of Wansink discussing his research at Chinese buffets that appears to contradict the methods described in the paper.)

Obesity editor Eric Ravussin told us the correction was:

…in response to a whistleblower which was specifically mentioning some problem with the table. We went to the author.

He noted the journal “investigated it as much as we can.”

In April, the editor of Evolutionary Psychological Science, Todd Shackelford, added an “editorial note” to one of the four papers initially questioned last year:

In the article “Eating Heavily: Men Eat More in the Company of Women,” by Kevin M. Kniffin, Ozge Sigirci, and Brian Wansink (Evolutionary Psychological Science, 2016, Vol. 2, No. 1, pp. 38–46), [the] authors report that the units of measurement for pizza and salad consumption were self-reported in response to a basic prompt “how many pieces of pizza did you eat?” and, for salad, a 13-point continuous rating scale. The authors further report that a robustness check that affirms the main findings that men tend to eat (i) more pizza and (ii) more salad in the company of women is now available at the CISER Data Archive at https://doi.org/10.6077/J5CISER2782, with raw data as well as Stata command lines that reproduce the signature results of the article.

The journal that published “Eating Heavily: Men Eat More in the Company of Women” is not indexed.

The note has also drawn criticism from outside researchers, including Anaya and Nick Brown, who argue the paper contains additional problems:

In view of these problems, we believe that the only reasonable course of action in this case is to retract the article, and to invite the authors, if they wish, to submit a new manuscript with an accurate description of the methods used, including a discussion of the consequences of their use of self-report measures for the validity of their study.

Anaya, Brown, and Tim van der Zee have just published their criticism of Wansink’s four pizza papers in BMC Nutrition; the original paper was posted as a preprint in PeerJ.

Like Retraction Watch? Consider making a tax-deductible contribution to support our growth. You can also follow us on Twitter, like us on Facebook, add us to your RSS reader, sign up on our homepage for an email every time there’s a new post, or subscribe to our daily digest. Click here to review our Comments Policy. For a sneak peek at what we’re working on, click here.

  • Post a comment

    Threaded commenting powered by interconnect/it code.