B1567
Title: Statistical properties of Cohen's d from linear regression
Authors: Juergen Gross - University of Hildesheim (Germany) [presenting]
Annette Moeller - Bielefeld University (Germany)
Abstract: The size of the effect of the difference in two groups with respect to a variable of interest may be estimated by the classical Cohen's d. A recently introduced generalized estimator allows conditioning on further independent variables within the framework of a linear regression model. The estimator may be derived by applying the so-called Frisch-Waugh-Lovell theorem in a partitioned linear regression model, thereby revealing similarities and required adjustments of classical formulas. Under normality assumptions, it is possible to derive distributional properties to compute standard errors and confidence intervals. The results fit between the classical effect size measure for the unconditional difference in two groups and the effect size measure f2, usually considered within an even more general regression context. The actual application of the findings can be illustrated with a publicly available dataset.
