Browse Journals

At present, there are three simple ways to describe the association of between participants binary variables. One of them is to use the phi coefficient. Another is to use the odds ratio or similar method (e.g., risk ratio or absolute risk reduction), possibly supplemented by Yule’s Q. A third method is to use the difference between conditional proportions. I propose another simple alternative, the adjusted success rate, not to replace these methods but as an addition to the researcher’s toolbox. I also compare and contrast the three methods with the adjusted success rate.

adjusted success rate, negative adjusted success rate, success rate, phi coefficient, difference between conditional proportions, odds ratio, Yule’s Q.