Applied Logistic Regression Analysis

by Scott Menard

• ISBN13: 9780761922087
• ISBN10: 0761922083
• Edition: 2nd
• Format: Paperback
• Copyright: 2001-10-09
• Publisher: Sage Publications, Inc

Summary

The focus in this Second Edition is on logistic regression models for individual level (but aggregate or grouped) data. Multiple cases for each possible combination of values of the predictors are considered in detail and examples using SAS and SPSS included. New to this edition:· More detailed consideration of grouped as opposed to casewise data throughout the book· Updated discussion of the properties and appropriate use of goodness of fit measures, R2 analogues, and indices of predictive efficiency· Discussion of the misuse of odds ratios to represent risk ratios, and of overdispersion and underdispersion for grouped data· Updated coverage of unordered and ordered polytomous logistic regression models.

Table of Contents

Series Editor's Introduction	v
Author's Introduction to the Second Edition	vii
Linear Regression and the Logistic Regression Model	1	(16)
Regression Assumptions	4	(7)
Nonlinear Relationships and Variable Transformations	11	(1)
Probabilities, Odds, Odds Ratios, and the Logit Transformation for Dichotomous Dependent Variables	12	(2)
Logistic Regression: A First Look	14	(3)
Summary Statistics for Evaluating the Logistic Regression Model	17	(24)
R2, F, and Sums of squared Errors	18	(2)
Goodness of Fit: GM, R2L, and the Log Likelihood	20	(7)
Predictive Efficiency: λp, τp, φp, and the Binomial Test	27	(9)
Examples: Assessing the Adequacy of Logistic Regression Models	36	(5)
Conclusion: Summary Measures for Evaluating the Logistic Regression Model	41	(1)
Interpreting the Logistic Regression Coefficients	41	(26)
Statistical Significance in Logistic Regression Analysis	43	(5)
Interpreting Unstandardized Logistic Regression Coefficients	48	(3)
Substantive Significance and Standardized Coefficients	51	(5)
Exponentiated Coefficients or Odds Ratios	56	(1)
More on Categorical Predictors: Contrasts and Interpretation	57	(4)
Interaction Effects	61	(2)
Stepwise Logistic Regression	63	(4)
An Introduction to Logistic Regression Diagnostics	67	(24)
Specification Error	67	(8)
Collinearity	75	(3)
Numerical Problems: Zero Cells and Complete Separation	78	(2)
Analysis of Residuals	80	(9)
Overdispersion and Underdispersion	89	(1)
A Suggested Protocol for Logistic Regression Diagnostics	90	(1)
Polytomous Logistic Regression and Alternatives to Logistic Regression	91	(12)
Polytomous Nominal Dependent Variables	94	(3)
Polytomous or Multinomial Ordinal Dependent Variables	97	(4)
Conclusion	101	(2)
Notes	103	(4)
Appendix: Probabilities	107	(1)
References	108	(3)
About the Author	111