UC Professor's Course on Statistical Fraud
"The announced opinions and published papers favoring mechanical fluoridation of public drinking water are especially rich in fallacies, improper design, invalid use of statistical methods, omissions of contrary data, and just plain muddleheadedness and hebetude. Many of the blunders were so glaring that I gave them to my beginning freshman classes in statistics at the very first meeting. The students see through them straightway, and are afforded great amusement. Uproarious laughter frequently ensues. No special statistical equipment is necessary to detect those peccancies. Of course the class and the Group soon tired of those infantilities, and sought and found greater challenge.
"By the way, a study by John Yiamouyiannis and Dean Burke on possible connection between cancer and waterborne fluoride was fairly tightly reasoned. The statistical procedures were standard, and much better applied than in much of the Public Health work.
"As I pointed out in a letter published in the proceedings of a congressional committee investigating the above connection, the real point is that direct chemical and controlled experimental research by unbiased uncommitted agencies is urgently indicated. Clearly fluoridation should be discontinued everywhere until definitive results on safety are obtained.
"In this connection, a great source of entertainment to the Group was the ferocity with which the researchers attacked any criticism. Invariably they violated in their own work the very principles they insisted on in others' work."
Hubert A. Arnold, PhD (1980)
Department of Mathematics,
University of California, Davis