Nutrition is messy. Even careful researchers disagree on diets, adherence and what transfers from lab to real life. Still, single papers are often packaged as hard "nutrition facts". The headline travels. The caveats do not.

Sometimes the problem is worse than sloppy communication. Data get stretched, hypotheses appear after the fact, or errors surface only under pressure. The Brian Wansink case at Cornell's Food and Brand Lab, alongside earlier fabrication scandals such as Diederik Stapel in the Netherlands, shows how fragile nutrition science becomes when reputation and media reach outrank scrutiny.

From fabricated clarity to sticker science

Stapel was dismissed in 2011 for inventing data and scrubbing inconvenient results. The pattern is familiar: unusually large effects, tidy stories, findings that feel too clean. Those are exactly the studies that spread.

A related failure mode played out in food psychology. A Cornell paper claimed that putting an Elmo sticker on an apple made children (reportedly aged 8 to 11) choose fruit more often at school lunch. It ran in JAMA Pediatrics and fit Wansink's public brand as the "Sherlock Holmes of food".

After criticism came a correction: misdescribed choice conditions, wrong sample sizes, unusual statistics. Wansink still insisted there were no further errors. Then the paper was fully retracted. The work had not studied 8 to 11-year-olds at all, but preschoolers aged 3 to 5. For parents and schools, that is not a footnote.

P-hacking: mine the data, invent the story

The deeper issue was process. Wansink collected buffet-price data in an Italian restaurant. His hypothesis (cheaper buffets mean less eating) did not hold. Instead of publishing the null, he urged a student to chase "plans B, C and D". Multiple papers soon emerged from the same dataset.

That is classic p-hacking or data dredging: fishing without a preregistered hypothesis until something looks "significant" (often p below 0.05). Dull or negative results disappear. Blog-friendly claims about pizza, regret and dining companions survive.

Researchers linked to Leiden and Groningen reviewed four of those "pizza studies" and reported roughly 150 inconsistencies in counts, stats and reporting. Not every inconsistency equals fraud, but that much noise makes the claims useless for coaching practice.

Why coaches and athletes should care

Sports nutrition advice often rides on studies that sound like newsletter gold: a sticker trick, a buffet trick, a simple hack. If the data are shaky, programs rest on sand.

  • Large effects from small behavioural samples deserve skepticism.
  • A paper that writes your headline for you is not evidence of quality.
  • Retractions and corrections belong at the top of your appraisal, not the bottom.
  • Null results are information. They vanish too often because they do not click.

For practical choices in healthy food and research, the hierarchy stays the same: repeated findings and systematic reviews outrank one viral lab story. The same bar applies across fitness training claims.

A faster filter for nutrition headlines

You do not need to be a statistician to raise better filters.

  • Check whether age group, setting and outcome actually match your athlete or client.
  • Ask whether the hypothesis was fixed in advance, or whether the same dataset was mined exploratively.
  • Look for independent replication. One lab or one press release is not consensus.
  • Ask what was not measured: calories, adherence, long-term effects, confounders.

Science does not need to be perfect to be useful. It does need to be honest about uncertainty. In a field flooded with marketing budgets, research reputation is one of the few real counterweights. Sticker-science claims do not belong there.

Read also

Bottom line

"Nutrition facts" are often stories with a p-value attached. Wansink and earlier fraud cases show how fast those stories can unravel. For FITsociety readers the takeaway is simple: save your trust for mechanisms that replicate, and keep sensational headlines at arm's length until the data hold up.

References

  1. RTL Nieuws coverage of Diederik Stapel dismissal for fabricated data.
  2. RTL Nieuws summary of key findings from the Stapel investigation.
  3. Wansink B, Just DR, Payne CR. Can Branding Improve School Lunches? Arch Pediatr Adolesc Med. 2012;166(10):967-968. doi:10.1001/archpediatrics.2012.999 (later retracted).
  4. Retraction Watch. Backlash prompts prominent nutrition researcher to reanalyze multiple papers (2017).
  5. Lee S. Reporting on the Elmo apple study age discrepancy. BuzzFeed News.
  6. JAMA Pediatrics retraction notice for the branding/school lunch paper.
  7. Just DR, Sigirci O, Wansink B. Lower buffet prices lead to less taste satisfaction. Journal of Sensory Studies. 2014;29(5):362-370.
  8. Just DR, Sigirci O, Wansink B. Peak-end pizza: Prices delay evaluations of quality. Journal of Product & Brand Management. 2015;24(7):770-778.
  9. Kniffin KM, Sigirci O, Wansink B. Eating heavily: Men eat more in the company of women. Evolutionary Psychological Science. 2016;2(1):38-46.
  10. Sigirci O, Wansink B. Low prices and high regret: how pricing influences regret at all-you-can-eat buffets. BMC Nutrition. 2015;1:36.
  11. van der Zee T, Anaya J, Brown NJL. Statistical heartburn: An attempt to digest four pizza publications from the Cornell Food and Brand Lab. PeerJ Preprints. 2017;5:e2748v1. https://doi.org/10.7287/peerj.preprints.2748v1