Handling Covariate Patterns in the Examination of Logistic Regression Residuals

• In Event: Linear, Nonlinear, Latent, and Machine Learning Modeling for Educational Research

Abstract

The Handling of Covariate Patterns in the Examination of Logistic Regression Residuals,
influence, and leverage, require residual diagnostics to be calculated by covariate pattern rather than by subject. Therein, a software requirement for correct analysis of many packages is to convert raw data into a compressed Event/Trial format. As transposing to that format manually is tedious and error prone, an Excel spreadsheet is offered that creates a copy and paste Event/Trial formatted data file automatically for that purpose. In addition, another Excel program is offered that correctly calculates residual diagnostic statistics via covariate pattern automatically internally from raw data, as well as all other usual logistic regression output. Examples are provided.

Authors

• Mary G. Lieberman, Florida Atlantic University

• John D. Morris, Florida Atlantic University

• Maria D. Vásquez-Colina, Florida Atlantic University