>"The FDA has done far, far more good than harm in this case."
How sure are you of that? I am certain that the organization is structured to make many more type II than type I errors, and I would guess that type II errors cause at least the same magnitude of harm and they are, by their nature, much harder to measure.
Comments
>"The FDA has done far, far more good than harm in this case."
How sure are you of that? I am certain that the organization is structured to make many more type II than type I errors, and I would guess that type II errors cause at least the same magnitude of harm and they are, by their nature, much harder to measure.
Random paper on the topic: http://repositories.cdlib.org/pep/96-2/