I was delighted to see a couple recent F1000 evaluations that strayed from traditional peer-reviewed literature. F1000 Members Frank Harrell, a biostatistician at Vanderbilt Medical School, and Daniel Beard, a bioengineer at the Medical College of Wisconsin, independently evaluated an article by Jonah Lehrer in The New Yorker magazine, “The truth wears off: Is there something wrong with the scientific method?” (Dec. 13, 2010). Rated “exceptional” and “must read” respectively, the article claims that many, many scientific findings are ultimately not replicable, and attributes this to individual biases brought to experimental and epidemiological research, as well as to the inherent failures of models and statistics. Harrell writes in his evaluation that “clinicians should be far more skeptical of medical journal articles,” and Beard warns that “the real scientific failure occurs when one substitutes statistical for scientific thinking.”
Cell biologist and bioethicist Fred Grinnell reflects on just these issues in his ...
