Learn R Programming

MOTE (version 1.2.2)

d_prop: Cohen's d (SMD) for Independent Proportions (Binary Outcomes)

Description

This function computes a standardized mean difference effect size for two independent proportions by treating each as the mean of a Bernoulli (0/1) variable and computing a standardized mean difference (SMD) directly using the pooled Bernoulli standard deviation. This follows the same logic as Cohen's d for continuous variables, but applied to binary outcomes:

Usage

d_prop(p1, p2, n1, n2, a = 0.05)

d.prop(p1, p2, n1, n2, a = 0.05)

Value

A list with the same structure as [d_ind_t()], containing the standardized mean difference and its confidence interval, along with auxiliary statistics. The list is augmented with explicit entries `p1`, `p2`, `p1_value`, and `p2_value` to emphasize that the original inputs were proportions.

Arguments

p1

Proportion for group one (between 0 and 1).

p2

Proportion for group two (between 0 and 1).

n1

Sample size for group one.

n2

Sample size for group two.

a

Significance level used for confidence intervals. Defaults to 0.05.

Details

$$d = \frac{p_1 - p_2}{s_{\mathrm{pooled}}}$$

where

$$s_{\mathrm{pooled}} = \sqrt{\frac{(n_1 - 1)p_1(1 - p_1) + (n_2 - 1)p_2(1 - p_2)} {n_1 + n_2 - 2}}$$

This replaces the original z‐based formulation used in older versions of MOTE. The SMD effect size is directly comparable to all other d‐type effect sizes in the package.

Examples

Run this code
d_prop(p1 = .25, p2 = .35, n1 = 100, n2 = 100, a = .05)

Run the code above in your browser using DataLab