csdvsmdd.fisher {MDD} | R Documentation |
For a trial evaluated using Fisher's exact test, calculates the probabilities of either having a clinically significant difference declared statistically insignificant, or of having a statistically significant difference declared clinically insignificant. Also ouputs the power of the test.
csdvsmdd.fisher(n.placebo, n.treat, prob.placebo, prob.treat, csd, alpha = 0.05, alternative = "two.sided")
n.placebo |
number of subjects in control group. |
n.treat |
number of subjects in treatment group. |
prob.placebo |
anticipated proportion of responses in control group. |
prob.treat |
anticipated proportion of responses in treatment group. |
csd |
clinically significant difference. |
alpha |
significance level. |
alternative |
indicates the alternative hypothesis and must be one of "two.sided" ,
"greater" , or "less" . You can specify just the initial letter. |
The “clinically significant difference” is the smallest difference that will be meaningful either clinically or commercially. This is usually obtained from the clinical scientists and/or business partners you are working with.
A list with the following three components:
statsig.not.clinsig |
the probability that the trial will have a statistically significant but clinically insignificant result. |
clinsig.not.statsig |
the probability that the trial will have a clinically significant but statistically insignificant result. |
power |
the power of the test. |
Don Barkauskas (barkda@wald.ucdavis.edu)
See fisher.test
for references.
csdvsmdd.chisq
for an asymptotic test
gui.csdvsmdd
for a GUI version
csdvsmdd.fisher(36, 36, .25, .6, .25) #Note that both possibilities are represented here. csdvsmdd.fisher(20, 20, .25, .6, .25) csdvsmdd.fisher(20, 20, .25, .6, .45) #Compare to the same commands using csdvsmdd.chisq; the first leads to identical #results, while the second and third give different results.