Ordinal Regression Analysis: Using Generalized Ordinal Logistic Regression Models to Estimate Educational Data

• In Event: Mediation, Ordinal, and Nonlinear Modeling

Abstract

The proportional odds (PO) assumption for ordinal regression analysis is often violated, since it is strongly affected by sample size and the number of covariate patterns. To deal with this issue, the partial proportional odds (PPO) model and the generalized ordinal logit model were developed. However, the use of these models seems to be overlooked. One major reason is the restrictions of the current statistical software packages: SPSS cannot perform the generalized ordinal logit model analysis, and SAS needs data restructuring. The purpose of this paper is to illustrate the use of generalized ordinal logistic regression models to predict mathematics proficiency levels using Stata, and compare the results of fitting the PO models and the generalized ordinal logistic regression models.

Authors

• Xing Liu, Eastern Connecticut State University

• Hari P. Koirala, Eastern Connecticut State University