On Thu, 16 Jan 2003 12:00:15 -0500, Rich Ulrich <[EMAIL PROTECTED]> wrote:
>On Thu, 16 Jan 2003 10:28:02 +0100, Michael Specka <[EMAIL PROTECTED]> >wrote: >> Agreed. But there might be circumstances where even small >> correlations provide important information and justify certain >> actions. If the 0.06 correlation comes from a death rate of >> 250 out of 500 in one treatment and 220 out of 500 in the other >> treatment... >> >As a clinician, you take note that a 15% advantage is >minuscule and not statistically significant, and hardly any >reason, by itself, to select between the two treatments. >- if it is "just-barely-significant", you can't show that it >is robust against contingencies; and if it is not-significant, >your confidence limit includes the opposite *conclusion*. I agree that small effects could often lack robustness against contigencies. On the other hand, if it should be necessary to decide between two actions, wouldn't it be sensible to base this decision on the data at hand? I do not mean to blindly assume that a p=0.05 is a 95% protection against deciding wrongly. And costs, consequences and context of the decision should be taken into account as far as possible. But basically I'd rather assume that action 1 is preferable to action 2, even if it it was only 15% superior in a given trial with n=500. Regards Michael Specka . . ================================================================= Instructions for joining and leaving this list, remarks about the problem of INAPPROPRIATE MESSAGES, and archives are available at: . http://jse.stat.ncsu.edu/ . =================================================================
