This program estimates statistical power for moderated multiple regression (MMR) using a theory-based approximation. It estimates statistical power to detect a dichotomous moderator variable with MMR using a theory-based approximation. It can be used for moderators with more than 2 levels and it also allows you to specify variable reliabilities as well as range restriction, making it the most comprehensive, and therefore recommended, power estimator.
INSTRUCTIONS: First, specify the number of sub-groups (i.e., moderator-based categories), desired a priori Type I error rate (e.g., .05., .01), and whether there are sampling restrictions (i.e., range restriction). Then, in the second screen, the program prompts you to input specific information based on your research scenario for each moderator-based sub-group: sample size, correlation between predictor X and outcome Y, standard deviations and reliability estimates for X and Y, and variance multiplying factor for X (i.e., expected sample variance of X divided by population variance of X). The third and final screen displays the same information you entered (so you can check that all the information you entered is accurate) and the power estimate. The program allows you to change values for, for example, sample sizes, reliabilities, and range restriction on X and shows you in real time how statistical power is affected by these research design and measurement characteristics.
Source: Aguinis, H., Boik, R.J., & Pierce, C.A. (2001). A generalized solution for approximating the power to detect effects of categorical moderator variables using multiple regression. Organizational Research Methods, 4, 291-323. [copyright notice: You may download this article for one-time personal use only; please obtain publisher permission for any further distribution, publication, or commercial use.][pdf]
