Simon, Steve, PhD wrote:
> There's another issue that you didn't ask about, but I'll tell you anyway
> (that's the price you have to pay for asking for free advice). Calculating
> power after the data is collected is a controversial area. Most post hoc
> power calculations are uninformative, since they are inversely related to
> the p-value. You can do okay if you remember to use a clinically relevant
> difference in the calculations rather than the observed difference.
Absolutely! I would add that post hoc power analysis can make sense if you
have an independent estimate of effect size (e.g. from previous research).
However, in this case power analysis is rather limited. If you have a good
idea of effect size it might beg the question of why you are doing the
research (replication is not a complete answer here because replication is
most useful when the effect size is still uncertain). For applied research it
is sometimes possible to calculate what a useful (e.g. a clinically relevant)
effect size would be - as you propose. This is often a better criteria than an
independent estimate of effect size (which is often likely to involve a
different study design or type of sample).
Thom
===========================================================================
This list is open to everyone. Occasionally, people lacking respect
for other members of the list send messages that are inappropriate
or unrelated to the list's discussion topics. Please just delete the
offensive email.
For information concerning the list, please see the following web page:
http://jse.stat.ncsu.edu/
===========================================================================
