Rick Adams
> Stephen Black wrote:
> > Bingo, double-blind study. QED.
>
> Except that, since the experimenters themselves would
> know if they were administering the placebo therapy or not,
> their interactions with the subjects could be compromised
> by the knowledge and the results skewed. :(
Remember that none of this happens in an epistemological vacuum. We want to
know whether or not a particular therapy is effective. So far we have
nothing to go on but the enthusiasm and clinical experience of the therapy's
advocates. Along comes Stephen with his almost-double-blind-but-still-flawed
study. So we have one study that removes several (but not all) sources of
potential error, and another source of evidence that fails to remove even
those sources of potential error.
Sounds like a study well worth doing, to me. If someone else comes along
and solves that remaining flaw (e.g, by use of graduate students to
administer the therapy), I'm swayed that way. But until then, Stephen's
study sounds like a valuable source of information.
(I know, the thread is about the challenge of designing true double-blind
tests of therapy...but it's alway worth remembering that science doesn't
work by producing a single, perfect test guaranteeing its results. It works
by pruning potential sources of error, one and two at a time, giving us
better and better - but eternally tentative - results. And that's plenty
good enough - far better than any alternatives).
Paul Smith
Alverno College
Milwaukee
