nPM: Estimation of Sample Size via Power and Mean

In Dcroix/RSamplingz: Package for Sampling Needs

Estimation of Sample Size via Power and Mean

Description

A formula to estimate the sample size using Power and Mean; preferred when there exist literatures that provide the needed statistics for its computation.

Usage

nPM(alpha, power, sd, delta)

Arguments

alpha	value set for Type I error
power	preferred power; indicated as 1 minus Type II error
sd	standard deviation
delta	the difference between the means of group A and B

Details

Computes for the sample size using the confidence level, delta, standard deviation and preferred power.

Value

Returns a character vector which presents the number of required samples for each group in the study.

Note

Unlike other formulas included in the package like Cochran and Smean, set alpha and power is converted to z-score by the nPM function

Author(s)

Paolo G. Hilado

Examples

nPM(0.05, 0.8, 50, 10)
[1] "393 each group"

## The function is currently defined as
nPM <- function(alpha,power,sd,delta){
a <- qnorm(1-(alpha/2))
p <- qnorm(power)
x <- a + p
x <- x*x
b <- sd*sd
c <- 2*x*b
n <- c / (delta*delta)
n <- ceiling(n)
ans <- paste(n, "each group")
print(ans)
}

Dcroix/RSamplingz documentation built on April 30, 2022, 8:33 a.m.