I want to know if my study and my statistical test is powerful.
I have 16 subjects, all of them took both treatment1 and treatment2. I want to know if the study is powerful by conducting a power analysis to determine the power of the study.
Group 1 : n=16, mean +/ sd => 0.9 +/- 1.4 ; Group 2 : n=16, mean +/ sd => 1 +/- 0.01
The two groups are paired (same subjects)
I found the package MKpower and function sim.ssize.wilcox.test but I don't quiet understand how to use. I tried this
>> rxy <- function(n) rnorm(n, mean = mean(group1),sd = sd(group1))
>> sim.ssize.wilcox.test(rx = rxy, n.min=8,n.max = 30,step.size = 2, iter = 10000,type="paired")
n emp.power
8 8 0.3181
10 10 0.4619
12 12 0.5253
14 14 0.6351
16 16 0.6866
18 18 0.7556
20 20 0.8049
Am I using this right? and if I follow the result it means the power for n=16 is 0.68, which seems hard to believe ...
which R function can calculate the power for Wilcoxon signed-rank test, please ?