ss_best_normal: Sample Size for Selecting the Best Treatment in a Normal Response (Indifference-Zone)
Description
Calculates the minimum common sample size per group needed to achieve a specified
probability (power) of correctly selecting the best group using the
indifference-zone approach. This method assumes normally distributed responses with
a known and common standard deviation.
Integer. Sample size required per group to achieve the specified power.
Arguments
power
Numeric. Desired probability of correctly selecting the best group.
dif
Numeric. Indifferent-zone. Minimum difference that is considered meaningful.
sd
Numeric. Common standard deviation of the response variable.
ngroups
Integer. Number of groups (treatments) being compared.
seed
Optional. Integer seed to use in the internal call to multz().
Details
The indifference-zone approach guarantees that the probability of correct selection
is at least power, assuming the best group's mean exceeds the others by at
least dif. The calculation is based on Bechhofer's Procedure Nb.
References
Bechhofer, R.E., Santner, T.J., & Goldsman, D.M. (1995).
Design and Analysis of Experiments for Statistical Selection, Screening, and Multiple Comparisons.
Wiley Series in Probability and Statistics. ISBN: 0-471-57427-9.