No label defined (Q4947973)

This book is a good introduction into methods and models for the analysis of categorical data and their applications in social sciences research. The aim of the book is to help applied social scientists to use special tools that are well suited for analyzing categorical data. The primary audiences are graduate students and practicing researchers in social sciences. One of the main aims of the book is to integrate the transformational approach and the latent variables approach, two diverse but complementary traditions dealing with the analysis of categorical data. The statistical, or transformational, approach to categorical data analysis is most familiar to researchers in demography and biostatistics, whereas the latent variables approach is often taken by economists. The authors set up a website for the book at: \url{http://www.la.utexas.edu/research/faculty/dpowers/book}. This website contains the data sets and programming codes of several popular statistical packages (e.g., GLIM, LIMDEP, SAS, STATA, and TDA) for the examples discussed in the book. The book is appropriate for a single-term course on categorical data modeling. Chapter 1, ``Introduction'', and Chapter 2, ``Review of linear regression models'', provide an introduction and basic foundation for the course. The regression-type modeling approach is considered as an appropriate analytic method. Chapter 3, ``Logit and probit models for binary data'', provides an introduction and detailed treatment of regression models for binary data. Chapter 4, ``Loglinear models for contingency tables'', goes into greater detail on the methods for analyzing contingency tables. Chapter 5, ``Log-rate models'', discusses models for transition rates. Chapter 6, ``Models for ordinal dependent variables'', and Chapter 7, ``Models for unordered dependent variables'', provide an overview of methods for ordered and non-ordered categorical responses.