Sample size for a standard deviation confidence interval

Computes the sample size required to estimate a population standard deviation with desired confidence interval precision. This function assumes that the traditional confidence interval for a population standard deviation will be used. The traditional confidence interval assumes that the response variable has an approximate normal distribution and can be highly innacurate when this assumption is not satisfied.

Usage

size.ci.sd(alpha, r)

Arguments

alpha: alpha level for 1-alpha confidence
r: desired upper to lower confidence interval endpoint ratio

Value

Returns the required sample size

Examples

size.ci.sd(.05, 1.5)
#>  Sample size
#>           49

# Should return:
# Sample size
#          49