I have 2 treatment groups (1st=control, 2nd=treated,
n=8 in each) for which we have parasite count data.
To summarise for each trt group, we use geometric
means, and calculate the confidence interval about the
mean as usual.
We also calculate a % reduction in parasite counts
from the control trt group to the treated group, using
100 * [geomean(treated) - geomean(control)] /
geomean(control)
(where geomean is the geometric mean)
How should I be calculating the confidence interval
around this % reduction?
Thanks in advance,
Fiona.
___________________________________________________________
Yahoo! Photos – NEW, now offering a quality print service from just 8p a photo http://uk.photos.yahoo.com
|