Skip to contents

Sample Size or Power Calculation for K Means

Source: R/k_mean_size.R
k_mean_size.Rd

Calculates sample size or power for a multiple-sample mean test.

Usage

k_mean_size(
  muA,
  muB,
  kappa = 1,
  sd = NULL,
  sdA = NULL,
  sdB = NULL,
  tau = 1,
  alpha,
  beta = NULL,
  n = NULL,
  nA = NULL,
  test_type = "2-side"
)

Arguments

muA

Numeric. True mean of group A.

muB

Numeric. True mean of group B.

kappa

Numeric. Ratio of sample sizes (nA/nB). Default is 1.

sd

Numeric (optional). Standard deviation. Required for "2-side" test.

sdA

Numeric (optional). Standard deviation of group A. Required for "1-side" test.

sdB

Numeric (optional). Standard deviation of group B. Required for "1-side" test.

tau

Integer. Number of comparisons.

alpha

Numeric. Type I error rate.

beta

Numeric (optional). Type II error rate (1 - power). Required for sample size calculation.

n

Integer (optional). Sample size. Required for power calculation of "2-side" test.

nA

Integer (optional). Sample size of group A. Required for power calculation of "1-side" test.

test_type

Character. "2-side" or "1-side". Default is "2-side"

Value

Numeric. Returns sample size (if beta is given), or power (if n/nA is given).

Note

Only one of beta (for sample size calculation) or n/nA (for power calculation) should be specified.

Required arguments by test_type:

  • "2-side":

    • For sample size: muA, muB, sd, tau, alpha, beta

    • For power: muA, muB, sd, tau, alpha, n

-"1-side":

  • For sample size: muA, muB, sdA, sdB, tau, alpha, beta

  • For power: muA, muB, sdA, sdB, tau, alpha, nA

Examples

# Sample size for `"2-side"` test
k_mean_size(muA = 5, muB = 10, sd = 10, tau = 1,
            alpha = 0.05, beta = 0.2, test_type = "2-side")
#> [1] 63

# Power of `"2-side"` test
k_mean_size(muA = 5, muB = 10, sd = 10, tau = 1,
            alpha = 0.05, n = 63, test_type = "2-side")
#> [1] 0.8013024

# Sample size for `"1-side"` test
k_mean_size(muA = 132.86, muB = 127.44, kappa = 2, sdA = 15.34, sdB = 18.23, tau = 1,
            alpha = 0.05, beta = 0.2, test_type = "1-side")
#> [1] 85

# Power of `"1-side"` test
k_mean_size(muA = 132.86, muB = 127.44, kappa = 2, sdA = 15.34, sdB = 18.23, tau = 1,
            alpha = 0.05, nA = 85, test_type = "1-side")
#> [1] 0.8020669