Multiple Regression with Discrete Dependent Variables

Multiple Regression with Discrete Dependent Variables

John G. Orme and Terri Combs-Orme

Print publication date: 2009

ISBN: 9780195329452

Publisher: Oxford University Press


This book presents detailed discussions of regression models that are appropriate for discrete dependent variables, including dichotomous, polychotomous, ordered, and count variables. The major challenge in using such analyses lies in the nonlinear relationships between the independent and the dependent variables, which requires the use of link functions, and particularly in interpreting and presenting the findings in ways that are clear and meaningful. Clear and simple language guides the reader briefly through each step of the analysis and presentation of results to enhance understanding of the link function, the key to understanding these nonlinear relationships. Throughout the book provides detailed examples based on the data, and readers may work through these examples by accessing the data and output on the Internet at the companion Web site. In addition, each chapter provides a list of recommended additional readings and Internet content.