David B. Wilson, PhD, George Mason University
Version date: 2023.11.27
Binary Dependent Variable: Proportions within Each Condition
Notes: This method approximates Cohen’s d and Hedges’ g from binary or dichotomous outcome data using the proportion (a value between 0 and 1) of successes (or failures) in each condition. Using failures instead of successes changes the direction of the effect. Three methods of computing this approximation are provided, the logit method, the Cox logit method, and the probit method. All three should provide roughly similar results except when the proportion of “yes” or “no” frequencies is small. If a proportion equals 0, the proportion of successes is set to .5/n, where n is the sample size for that condition.
This work is licensed under a Creative Commons Attribution-Noncommercial 3.0 United States License.