'Lying Dutchman' Could Cast Doubt on Ad Research
Do women feel bad about themselves when they see ads for beauty and fashion products?
We'll probably never know for sure, because an investigation has confirmed the Dutch professor of psychology who did a study on the topic has falsified research in at least 30 scientific papers. The Washington Post called him "the Lying Dutchman." But it is still undetermined whether the ad study on beauty was also fraudulent until a final report is released by a committee investigating the matter.
The scope of the fraud is "considerable and shocking," stated the committee set up to look into the transgressions of Tilburg University professor Diederik Stapel.
"I have failed as a scientist and researcher," Mr. Stapel wrote on the Tilburg website. "I feel ashamed for it and have great regret."
Published in the University of Chicago Press' Journal of Consumer Research, Mr. Stapel's article purported to show that women who saw fashion or beauty products in ads had lower self-esteem than women who viewed the same items in a nonadvertising setting. It is one of many being investigated by the committee created at Tilburg.
Mr. Stapel was suspended when doubts emerged about his research concluding that eating meat makes people antisocial and selfish, but the matter raises broader questions about research in the ad industry.
The fraud is a "tragedy," said Geoffrey Precourt, editor of the Advertising Research Foundation's Journal of Advertising Research. Mr. Stapel's paper was "nonsense," Mr. Precourt said. "Everything we know about consumer research seems to be contradictory of the findings."
The editor did not think incomplete or misleading research was a serious problem among the 50 or so legitimate advertising journals. "The peer-review process pretty well sorts out the bad eggs," he said. That Mr. Stapel was published repeatedly indicates flaws in the system, however, Mr. Precourt admitted. "It should be the mission of all the editors … to step up their review process to guard against future frauds." ARF has always required full access to data, "but now we're going to reinforce it -- put some speed bumps in the way."
It would have taken a lot of speed bumps to deter Mr. Stapel. The Tilburg committee stated that to its knowledge "misconduct of this kind by a full professor is unprecedented." It has done "great harm to science and the field of social psychology in particular."
If the panel concludes that Mr. Stapel's paper was based on fraudulent research, the publication will retract the article, said Ann L. McGill, one of the Journal's editors. It will wait for Tilburg's final report, she said. Another ad paper by Mr. Stapel has already been removed from the publication's website, she said.
The fraud involved data created from nonexistent subjects. In the ad study, Mr. Stapel claimed to have conducted four experiments with 583 female students who earned course credits for participating. He described the tests in detail, complete with mathematical formulas and jargon.
The ad paper was a dissertation by Debra Trampe, an assistant marketing professor at the University of Groningen in the Netherlands under Mr. Stapel's supervision. A third author also contributed. The committee said it had "found no signs that the co-authors knowingly cooperated with Mr. Stapel in tampering with data," adding that he had misled students and co-authors "with great subtlety."
Ms. Trampe said that the inquiry would determine if Mr. Stapel falsified her research data. "That is why I am unable to comment on specific papers."
The committee described Mr. Stapel's scam as a "cunningly simple system" of fabrication. He would develop "intensive" partnerships with other scientists to develop plausible theories and lend credibility to his projects. "Up to this point everything was in order," the panel said. "There then followed an entirely fictitious phase." Mr. Stapel sometimes even invented research assistants who supposedly reported to him, and he kept his partners out of the loop.
The editors of publications in which Mr. Stapel's papers appeared took a lot on faith, the committee said. It is not a "luxury for journals to demand a detailed report of the research procedures followed to be made available on the internet," it asserted.
Robert Barocci, ARF's president-CEO, said he participates in peer reviews of material submitted to the Journal. No one has ever questioned its procedures, he said, "and we've been around for 50 years." He acknowledged that his judgment of articles is based on "substance, content and application" rather than on research methodology and data specifics.
Ms. McGill, a professor of marketing and behavioral science at the University of Chicago's Booth School of Business, said that "multiple-study papers" such as Mr. Stapel's are typically easier to confirm. But "if someone wanted to [enter] false data, I don't know that we could catch that ," she added. "We're not looking for fraud; we're looking for ways science can go awry."
The question for readers of these articles, of course, is : "Should I risk big money by incorporating this information into my marketing plan?"
Marketers and industry researchers point out that it is unlikely that any company would base decisions on one study in an academic journal -- especially a piece that makes such serious claims. "I can't imagine a big company" making such a move, said a marketing research director at a major packaged-goods company. "You'd want to do primary research on your own."
Mr. Barocci ventured that Mr. Stapel "isn't the first guy to abuse the system." Academicians are "intent to publish stuff," he said. "If [the work] can get into journals, readers assume … they can trust the material. Their trust is what has been violated.
"Just because it's science doesn't mean it's science," Mr. Barocci said.