Summary
Contents
Logit Modeling represents a breakthrough for researchers because it offers ways for more efficient estimation of models with multiple categorical variables, particularly whenever the measurement assumptions for classical multiple regression fail to be met. Taking an applied approach, DeMaris begins by describing the logit model in the context of the general loglinear model, moving its application from two-way to multidimensional tables. He then divides the rest of the book between an examination of the varieties of logit models for contingency tables and logistic regression. Throughout his coverage of both these major areas, DeMaris emphasizes interpretation of results. The book concludes with an extension of logistic regression to dependent variables with more than two categories.
Introduction
Introduction
Not too long ago, a variety of disparate techniques were needed to model a categorical dependent variable as a function of a set of explanatory variables. These techniques ranged from the repetitious elaboration of two-way cross-tabulations at fixed levels of the other predictors, to ordinary least squares (OLS) regression or discriminant analysis for a binary dependent variable, to discriminant analysis for a polytomous dependent variable ...