ht_2pop_prop {statBasics}

Hypothesis testing for two population porportions

Description

Comparing proportions in two populations

Usage

ht_2pop_prop(
  x,
  y,
  n_x = NULL,
  n_y = NULL,
  delta = 0,
  alternative = "two.sided",
  conf_level = NULL,
  sig_level = 0.05,
  na_rm = FALSE
)

Arguments

x

a vector of 0 and 1, or a scalar of count of sucesses in the first group.

y

a vector of 0 and 1, or a scalar of count of sucesses in the first group.

n_x

a scalar of number of trials in the first group.

n_y

a scalar of number of trials in the second group.

delta

a scalar value indicating the difference in proportions (\Delta_0). Default value is 0.

alternative

a character string specifying the alternative hypothesis, must be one of "two.sided" (default), "greater" or "less".

conf_level

a number indicating the confidence level to compute the confidence interval. If conf_level = NULL, then the confidence interval is not included in the output. Default value is NULL.

sig_level

a number indicating the significance level to use in the General Procedure for Hypothesis Testing.

na_rm

a logical value indicating whether NA values should be removed before the computation proceeds. Default value is FALSE.

Details

ht_2pop_prop can be used for testing the null hipothesis that proportions (probabilities of success) in two groups are the same.

If is.null(n_x) == T and is.null(n_y) == T, then x and y must be a numeric value of 0 and 1 and the proportions are computed using x and y. If is.null(n_x) == F and is.null(n_y) == F, then x, y, n_x and n_y must be non-negative integer scalars and x <= n_x and y <= n_y.

Value

a tibble with the following columns:

statistic

the value of the test statistic.

p_value

the p-value for the test.

critical_value

critical value in the General Procedure for Hypothesis Testing.

critical_region

critical region in the General Procedure for Hypothesis Testing.

delta

a scalar value indicating the value of delta.

alternative

character string giving the direction of the alternative hypothesis.

lower_ci

lower bound of the confidence interval. It is presented only if !is.null(conf_level).

upper_ci

upper bound of the confidence interval. It is presented only if !is.null(conf_level).

Examples

x <- 3
n_x <- 100
y <- 50
n_y <- 333
ht_2pop_prop(x, y, n_x, n_y)

x <- rbinom(100, 1, 0.75)
y <- rbinom(500, 1, 0.75)
ht_2pop_prop(x, y)