BAYESIAN STATISTICS AND DENTAL FLOSS. Fumento raises another serious objection to excluding evidence of the usefulness of flossing. Excluding everything but randomized tests excludes â€œbiological plausibility.â€ There isn’t much statistical evidence because “very few people have researched what is so very obvious: rotting food pressed against a mucous membrane or tooth enamel has got to be bad.”
In connection with the usefulness of breast self examination, I posted here on the major split between two schools of statisticians and wrote: â€œIn conducting a statistical test, one grouping (â€˜Bayesiansâ€™) wants to include independent knowledge and beliefs; the other wants to start from scratch (â€˜frequentistsâ€™).â€
Independent knowledge, which a Bayesian would use, would suggest that breast self examination might lead to earlier detection of breast cancer and that earlier detection might reduce mortality. That kind of independent evidence is excluded by frequentists.
Fumento quotes a dentist stating what a Bayesian might call “independent knowledgeâ€œ: “People daily build up a bio-film on their teeth called plaque, which is a layer of organic matter that contains some 500 bacterial species, food debris and other substances….â€ It seems plausible that breaking up plaque build up with floss would be useful, but that kind of independent evidence is excluded by frequentists.