I was wondering whether anyone could tell me the
effect, statistically, of applying an explicit mask
for a random fx analysis. I assume it is an anatomical
mask and so would mean a less stringent correction for
multiple comparisons.If this is so, is there any
effect on the uncorrected statistic, i.e. does it in
some way increase the statistical power overall. As a
general question is there any way of increasing the
power at the estimation stage.
Thanks
Micheal
__________________________________
Do you Yahoo!?
SBC Yahoo! DSL - Now only $29.95 per month!
http://sbc.yahoo.com
|