I have used the Kaplan-Meier test to analyse bond strengths for 3 different
adhesive systems with 30 bonds in each system, and subsequently a log-rank
test to compare the failure curves. I need to calculate the power of this
test before I can carry out further experiments. The only information I can
find on power / sample size calculations for this type of experiment is in
the medical literature and seems to relate to only two treatments and where
the end point is that a specified number of samples (patients) do not fail
which is not the case in this experiment as every bond fails.
I would be grateful for any advice or references on how to approach this
problem.
Many thanks.
Paul
_________________________________________________________________
Winks & nudges are here - download MSN Messenger 7.0 today!
http://messenger.msn.co.uk
|