Dear group members!

A trial will be stopped if a new sample size calculation shows that a
requested number of subjects is larger than N (because for more than N
subjects the trial will take too long). If the study is stopped, the
analysis will be done with a significance level of 0.05. If not, an
interim analysis will be done with an adjusted type I error (Pocock
alpha spending function with an overall type I error of 0.05).

Is this correct? Or is there also a need for an adjustment of type I
error if the trial will be stopped? If yes, how to adjust? And, has
there to be an adjustement in the interim analysis, too (in addition
to Pocock for the interim)?

My personal opinion, so far, is that the procedure described is okay,
because the sample size calculation only needs the standard deviation
and no mean at all; also it has nothing to do with a decision in terms
of rejecting or accepting the null hypotheses. But I am not sure at
all!

Hope, somebody can help!

Karl Knoblich
.
.
=================================================================
Instructions for joining and leaving this list, remarks about the
problem of INAPPROPRIATE MESSAGES, and archives are available at:
.                  http://jse.stat.ncsu.edu/                    .
=================================================================

Reply via email to