This book provides a comprehensive introduction to methods and models for categorical data analysis and their applications in social science research. An explicit aim of the book is to integrate the transformational and the latent variable approach, two diverse but complementary traditions dealing with the analysis of categorical data. This is the first introductory text to cover models and methods for discrete dependent variables, cross-classifications, and longitudinal data in a rigorous, yet accessible, manner in a single volume.The second edition of this book includes new material on multilevel models for categorical data. Several chapters have undergone extensive revisions and extensions to include new applications and examples. Highlights of the 2nd edition include a detailed discussion of classical and Bayesian estimation techniques for hierarchical/multilevel models, extensive coverage of discrete-time hazard models and Cox regression models, and methods for evaluating and accommodating departures from model assumptions. The accompanying website contains programming scripts to replicate each example using various statistical packages, which has proven to be an invaluable resource for instructors, students, and researchers.
This book presents the essential methods and models that form the core of contemporary social statistics. The book covers a remarkable range of models that have applications in sociology, demography, psychometrics, econometrics, political science, biostatistics, and other fields. It will be especially useful as a graduate textbook for students in advanced social statistics courses and as a reference book for applied researchers. Companion website also available, at https://webspace.utexas.edu/dpowers/www/
Industry Reviews
.,." the first introductory text to cover, in a single volume, models and methods for discrete dependent variables, cross-classifications, and longitudinal data. A great strength of the text is the authors' informal yet sophisticated approach, which combines the discussion of general principles with illuminating and realistic empirical examples." -Roberto Mare, University of California, Los Angeles, USA "Teaching this book will be almost too easy. The prose is clear, the examples are well chosen, and the Web site provides practical details." -Michael Hout, University of California, Berkeley, USA An excellent job done by the authors. As with the first edition, Powers and Xie make the analysis of categorical data easy to understand. There are 7 chapters that are clearly written, begining with a review of simple linear regression, then going to loglinear models for contingency tables, models for ordinal and nominal dependent variables and models for event ocurrence (models for rates). The technical level is high enough to understand the theory behind the analysis and the interpretation of results. The inclusion of a new chapeter (ch.5) on multilevel models is very clearly written, and includes a short introduction to modern Bayesian modelling. Although there is not enough space for a complete introduction into this topic (which requires a high level of mathematical statistics) the authors refer to other books (like the one by Scott Lynch) for more detailed explanations (needed for a better understanding) of bayesian modelling in general. This book is a great addition to the library of students and scientists in areas like biology and sociology who want an explained compendium of (modern) techniques for analysing categorical data. Amazon review
