Categorical Data Analysis Using The SAS® System, 2nd Edition

16.1. Introduction

Chapters 2–6 discuss methods for testing hypotheses of no association in two-way and stratified two-way contingency tables. These approaches focus on hypothesis testing rather than on model fitting and parameter estimation. In contrast, Chapters 8–12 describe logistic regression and weighted least squares methods for modeling a categorical response variable as a function of one or more categorical and/or continuous explanatory variables. These methods, which are analogous to ANOVA and regression techniques for normally distributed response variables, are appropriate when there is a clearly defined response variable of interest and you want to model how the response is affected by a set of explanatory variables or design factors. ...