When scientists discuss papers:
"I don't think this inference is entirely reasonable. If you're using several non-independent variables you're liable to accumulate more error than your model accounts for."
When scientists discuss grants:
"A guy who worked at the NSF once told me if we light a candle inside this jackal skull, the funders will smile upon our hopes."
"I'll get the altar!"
~ Zach Weiner, SMBC #2559
(1) Do people act more rationally when their interests are more directly concerned? (2) Are scientists' interests more directly concerned with winning grants than with making correct scientific inferences?
If the answer to both is "yes," then I think we should raise our confidence in jackal rituals relative to the current methodologies of statistical inference.
Here's the new thread for posting quotes, with the usual rules: