An Open Source R Shiny Web Application to Estimate Power for a Logistic Regression Interaction Term in the 2 X 2 X 2 Case

• Activity Number: 202
• Type: Contributed
• Date/Time: Monday, August 10, 2015 : 10:30 AM to 11:15 AM
• Sponsor: Biometrics Section
• Author(s): D. Keith Williams* and C. Heath Gauss and Zoran Bursac
• Companies: University of Arkansas for Medical Sciences and University of Arkansas for Medical Sciences and University of Tennessee Health Science Center
• Keywords: Power ; Sample Size ; Logistic Regression ; Interaction

Abstract

Researchers that are interested in measuring the difference in odds ratios between a pair of 2 x 2 tables are essentially investigating an interaction term of a logistic regression model. Presently there are few tools to power such an interaction effect, especially if the researcher is interested in utilizing the likelihood ratio test. Some empirical algorithms and associated SAS macros have been suggested and piloted by Bursac et. al. This work translates their algorithm into R code that is used to produce an open source R Shiny web application that allows widespread use.