Seth's Blog

• Experiments suggest flu shots reduce heart attacks and death. Huge reduction: 50%. The new report (a conference talk, not a paper) is a reanalysis of four earlier experiments. I was surprised to learn that the CDC uses heart attack outbreaks to locate flu outbreaks, implying that the new finding is not a fluke — there really is a strong connection. I already knew heart attacks are more common in the winter, which also supports a connection with flu.
• Une histoire des haines d’écrivains by Boquel Anne and Kern Etienne. Published 2009. About literary feuds. One of my students was reading a Chinese translation.
• Correspondences between sounds and tastes.
• Report on fraudulent Dutch research. “The 108-page report says colleagues who worked with Stapel had not been sufficiently critical. This was not deliberate fraud but ‘academic carelessness’, the report said.” I doubt it. Based on my experience with Chandra, I believe Stapel’s colleagues had doubts but did nothing from some combination of careerism (doing something would have cost too much, for example  a lot of time, and gained them nothing), ignorance (not their field), and decency (they saw no great value in ruining someone).  I wonder if the report considered these other possible explanations (careerism, ignorance, decency).

Thanks to Tim Beneke.

• Trying to crowd-source a cure for brain cancer.
• Interview with Aaron Blaisdell about Ancestral Health Symposiums
• Plagiarism Today, a website.
• Plagiarism averted
• Better late than never. By having him do a Fiction Podcast, the editors of The New Yorker finally acknowledge that David Sedaris is a fabulist. Treating his stories as if they actually happened (putting them under Personal History, not Fiction) was a curious editorial decision for a magazine that fact-checks poetry. Step 2 toward more New Yorker editorial honesty:  Book excerpts are labelled as such.
• self-experimentation workshop in Silicon Valley

Thanks to Dave Lull.

• You can major in Fermentation Science. No joke. When I was eight, I learned the concept of college major. I asked my mom, “What did you major in?” “Extracurricular activities,” she said. I failed to get the joke. She later explained she had spent more time working on the school paper than on her classes.
• In a famous paper, the statistician Ronald Fisher accused Mendel of faking his data. Fisher wrote: “the data of most, if not all, of the experiments have been falsified so as to agree closely with Mendel’s expectations.” This is not terribly consistent with the fact that Mendel’s highly improbable conclusions were correct. It’s as if Fisher had said “Person X used false info to claim he is worth $10 billion” and (b) in fact Person X is worth $10 billion. You can see that (a) and (b) may both be literally correct but that the term “false info” (Fisher’s “falsified”) probably conveys the wrong impression. This paper (“A Statistical Model to Explain the Mendel–Fisher Controversy”) has a more plausible explanation of the pattern in the data that Fisher noticed.
• Conflict of interest in the Nobel Prize in Literature. The conflicts of interest underlying the Nobel Prize in Physiology and Medicine — which are given out for “pure” science, thus  justifying more funding — remain unnoticed by journalists.

Thanks to Bryan Castañeda.

• BMJ video about overtreatment. Shannon Brownlee is one speaker.
• The myth of Jim Lehrer, neutral presidential debate moderator.
• Germany’s Education Minister, Annette Schavan, accused of plagiarism. “In previous cases, Schavan proved to be one of the most outspoken critics of those caught plagiarizing.” She is the second German Cabinet minister in the last year or so to face plagiarism charges. The first one resigned.

Thanks to Anne Weiss.

Richard Smith, former editor of the BMJ, writes about Peter Wilmshurst, a British cardiologist, an unusually brave and honest man:

He was the coprincipal investigator on a trial funded by NMT, an American company, to see whether closing a hole in the heart of patients with migraine would cure their migraine. It didn’t. He refused to agree to be an author on a paper published in the journal Circulation because the paper was misleading, and he gave an interview to a journalist in the US pointing out the problems in the study. NMT sued him for libel, not in the US, where proving libel is difficult, but in England, where the onus is on the defendant to prove his innocence.  NMT probably assumed (rightly in the case of most people) that the financial risk would cause Wilmshurst to cave in. They were wrong, and the case collapsed when NMT went bust.

The interesting thing about the stories Smith tells about Wilmshurst is the high rate of misconduct they imply:

[In 1996] we invited him to come to the BMJ and give a talk—behind closed doors—to our staff and advisers and colleagues from the Lancet. He reeled off case after case of misconduct, many of them involving prominent people. The audience listed intently, but I was unsure of the reaction. Might somebody leap up and say “How dare you accuse x of misconduct. He is one of the great men of British medicine”? In fact in my memory the reaction was the opposite. People said things like “Actually, it’s worse than you know…” . . .

Many of [Wilmshurst's stories] involve doctors who are guilty of misdemeanours but who sit in judgement on others. He told the story of Peter Richards who decided to bury the fact that Clive Handler, a doctor, at Northwick Park Hospital, was found guilty of using NHS research funds to subsidise his private practice at a time when Richards was medical director of the hospital and chair of the professional conduct committee of the GMC. Previously he had been dean of St Mary’s Medical School, prorector for medical education at Imperial College, and chair of the Council of Deans of UK Medical School and Faculties. When Handler eventually appeared before the GMC, the GMC’s lawyers ask that Richards stand down from chairing the committee. As Wilmshurst said, it’s as if a judge at the Old Bailey were to say “I’ll have to excuse myself from hearing this case as I helped the accused bury the body.” After having to stand down from this committee Richards continued to chair other conduct committees. Wilmshurst told several stories of doctors who had been found guilty of research misconduct but [had] gone on to be deans and others in charge of researchers.

Smith does not point out what this means. Doctor X is found guilty of research misconduct. Everyone knows this. Doctor X is still appointed dean or whatever. Maybe the people who make these appointments don’t care. Or maybe research misconduct is so common it cannot be a disqualification.

At the end of the article Smith points out his long friendship with Wilmshurst:

RS has known Wilmshurst for 16 years . . . the BMJ was sued for libel over an article by Wilmshurst that was published when RS was editor of the journal. The article has not been retracted but is not available on the BMJ website.

Presumably Smith was forced by BMJ lawyers to be this vague. I have not been able to locate the article. From  a talk by Wilmshurst. “Eventually Rubin got his report published in [The New England Journal of Medicine, under the editorship of Dr. Arnold Relman], because he threatened that unless his report was published he would go to the press and point out the collusion between the journal and Sterling-Winthrop.”

Wilmshurst’s conclusion: “Dishonesty is common in medical research.”

Thanks to dearime.

In a letter linked to by Nature, Daniel Kahneman told social psychologists that they should worry about the repeatability of what are called “social priming effects”. For example, after you see words associated with old age you walk more slowly. John Bargh of New York University is the most prominent researcher in the study of these effects. Many people first heard about them in Malcolm Gladwell’s  Blink. (more…)

• Propaganda in the Helping Professions by Eileen Gambrill, a professor in the School of Social Welfare at UC Berkeley. “Professionals can take courses in just about anything to satisfy continuing education requirements.” A book about false claims made by health care professionals.
• Fermented food causes fire trucks to show up
• Plagiarism at University of Waterloo. Why did they think they wouldn’t be caught?
• podcast about fermented foods

Terrence Deacon is a professor of anthropology at the University of California Berkeley — at the moment, chair of the anthropology department. (Deacon, like me, is interested in the evolution of language.) How unfortunate for the department, especially his graduate students, that he has recently been accused of using vast amounts of another person’s work without giving her credit. It isn’t easy to see the overlap, maybe because Deacon is a terrible writer (“by far the most unreadable book I have ever encountered” said a reviewer of one of his books), but there appears to be no doubt of the similarity and Deacon’s exposure to the work he is accused of not citing. Deacon says he doesn’t remember it.

When I brought unquestionable examples of plagiarism by Leslie Iversen, an Oxford professor, to the attention of Julie Maxton, the Registrar of Oxford University, she dismissed them (“honest error” — appearing to say that Iversen didn’t know that word-by-word copying is wrong). In this case there is no word-by-word copying but the failure to cite is far more upsetting to the persons not cited. What may have been copied is more abstract (“deeper”, you could say) and therefore more important.

At first, the complaint was dismissed. “I have concluded that the information available to me does not warrant appointment of an Investigative Officer under our campus faculty disciplinary procedures. The conduct you have alleged would not constitute a violation of the University of California’s Faculty Code of Conduct,” wrote Janet Broughton, Vice Provost, on May 27, 2011.

Later (December 12, 2011), Robert Price, Associate Vice Chancellor for Research, responded, ” The fact that certain concepts or phrases used by Dr. Deacon in the article you provided are the same or similar to concepts that appear in chapters from your book is not evidence of plagiarism, as these concepts may not be unique to your work. Perhaps the way you use these concepts is unique, which then would constitute plagiarism.” This understates the evidence, which is a long series of similarities.

Finally, the fact that outsiders find the claims (of failure to give credit) credible appears to have convinced Price that something must be done. “The continuing public dispute that your claims have generated lead me to believe that such an investigation [of the claims] is necessary in order to “clear the air”", wrote Price. After being successfully pressured to investigate, Price, a political science professor, in a September 3, 2012 letter reveals a lack of understanding of social pressure:

I wish to make crystal clear to you, your associates, and to all those to whom you are communicating that the University of California, Berkeley has not found that Professor Deacon has engaged in any form of research misconduct. The sole reason for undertaking an investigation are the claims made by you and your associates. [Contradicting what he said earlier -- that the "continuing public dispute" led to the investigation.] . . . The idea that you would use my communications with you ["use" in the sense of posting on a website] and the ongoing examination of your allegations by UC Berkeley as part of what increasing strikes me as a vendetta against Professor Deacon is reprehensible.

My letter to Alicia Juarrero [who complained] ends with this paragraph: “Our University policy on research misconduct, as well as the federal regulation on which it is based, require that all stages in the research misconduct investigation procedure are treated as strictly confidential. (UCB “Research Misconduct: Policies, Definitions and Procedures,” item IC and Federal Regulation 45CFR93.108). I expect that you will adhere to this requirement.” Rather than adhere to the stated requirement of confidentiality, Dr. Juarrero shared the letter with you and you, in turn, posted it on your website. What purpose is being served other than to make it appear that Deacon is guilty of something before even a single one of your claims has been validated? This sort of tactic will be familiar to those who remember the history of Joe McCarthy.

“What purpose is being served”? Uh, making the accusations harder to ignore? As for McCarthy, he made accusations without supplying evidence (“In this envelope [which he didn't open] I have the names of 80 Communists in the State Department”). That is not happening here.

Why Alicia Juarrero Got Mad at Terry Deacon. New Terry Deacon Website.

Someone named Red Fury made an interesting comment on my Boing Boing article about acne:

I had acne on/off for years. . . . In my mid-thirties, I tried the Retin-A at night, antibiotic gel for day regimen for about 2 years – no effect. . . . Then, I was talking to a co-worker whose daughter was taking ‘modeling classes’ to become a teen model. She casually mentioned her acned daughter had to give up rice, potato chips, and bread, all of which are high-glycemic index foods. My quack-radar went off, and I looked around for something scientific behind that advice. http://www.ajcn.org/content/86…

Huh. I guess those  nutrition-bashing dermatologists actually did a study and published the scientific results in a peer-reviewed journal. . . . My acne disappeared completely as soon as I eliminated rice and potatoes.

He finds a study that supports the casual advice, he follows the advice, his acne disappears. By convincing him to follow the advice, the scientific study helped him get rid of his acne. Which is impressive.

The interesting twist is that the study was published twice, clearly breaking the rules. Bad scientists! Who did something really good.

• Yale cancels China year-abroad program. One reason is that students in Beijing (at Peking University, one of the best universities in the country) learned less Chinese than students at Yale.
• European plagiarism epidemic. “More than a third of a new book for law students on how to write papers properly was plagiarised . . . The authors vowed to find the culprits.”
• Penkowa for Dummies. Complicated scientific fraud.
• Self-tracking difficulties: a fickle and too-demanding exercise tracker. I try to walk 60 minutes/day. It’s easy to track.

Thanks to Anne Weiss.