"Effect size" is a statistical term that means more or less what it would mean in ordinary speech: It is the size of the effect. How to measure effect size depends on the type of statistical analysis. For analysis of variance (ANOVA) and multiple regression, one measure of effect size is the coefficient of determination, usually labeled R^2. Another is f^2, which was invented by Jacob Cohen and is sometimes called Cohen's f^2.
R^2 is part of the computer output from every major statistical package (e.g. R, SAS, SPSS and many others) and can also be gotten from Excel or from some scientific calculators.
Instructions
-
-
1
Subtract R^2 from 1. For example, if R^2 = 0.3, 1 - 0.3 = 0.7.
-
2
Find the inverse of the result. This can be done on most calculators with a key often labeled 1/x. In the example 1/0.7 = 1.43.
-
-
3
Multiply the result by R^2. This is f^2. In the example, 1.43*0.3 = 0.43.
