Preface
Contents
1. Introduction
1.1. Categorical Response Data
1.2. Organization of This Book
Chapter Notes
Problems
2. Describing Two-Way Contingency Tables
2.1. Table Structure for Two Dimensions
2.2. Ways of Comparing Proportions
2.3. Summary Measures of Association
2.4 . Historical Overview
Chapter Notes
Problems
3. Inference for Two-Way Contingency Tables
3.1. Sampling Distributions
3.2. Testing Goodness-of-Fit
3.3. Testing Independence
3.4. Large-Sample Confidence Intervals
3.5. Exact Tests for Small Samples
3.6. Exact Non-null Inference
Chapter Notes
Problems
4. Models for Binary Response Variables
4.1. Generalized Linear Models
4.2. Logistic Regression
4.3. Logit Models for Categorical Data
4.4. Using Models to Improve Inferential Power
4.5. Probit and Extreme Value Models
4.6. Model Diagnostics
4.7. Fitting Logit Models
4.8. Conditional Logistic Regression
Chapter Notes
Problems
5. Loglinear Models
5.1. Loglinear Model for Two Dimensions
5.2. Table Structure for Three Dimensions
5.3. Loglinear Models for Three Dimensions
5.4. Loglinear Models for Higher Dimensions
Chapter Notes
Problems
6. Fitting Loglinear and Logit Models
6.1. Sufficiency and Likelihood for Loglinear Models
6.2. Estimating Expected Frequencies
6.3. Testing Goodness of Fit
6.4. Estimating Model Parameters
6.5. Iterative Maximum Likelihood Estimation
6.6. Analyzing Rates and Survival Times Using Loglinear Models
6.7. Table Standardization
Chapter Notes
Problems
7. Building and Applying Loglinear Models
7.1. Partitioning Chi-Squared to Compare Models
7.2. Strategies in Model Selection
7.3. Analysis of Residuals
7.4. Testing Conditional Independence
7.5. Estimating and Comparing Conditional Associations
7.6. Sample Size and Power Considerations
7.7. Empty Cells and Sparseness in Contingency Tables
Chapter Notes
Problems
8. Loglinear-Logit Models for Ordinal Variables
8.1. Linear-by-Linear Association
8.2. Row Effects and Column Effects Models
8.3. Models for Ordinal Variables in Multidimensional Tables
8.4. Testing Independence for Ordinal Classifications
8.5. Other Models Having Parameter Scores
8.6. Model Selection for Ordinal Variables
Chapter Notes
Problems
9. Multinomial Response Models
9.1. Generalized Logit Models and Loglinear Models
9.2. Multinomial Logit Models
9.3. Logits for Ordinal Responses
9.4. Cumulative Logit Models
9.5. Cumulative Link Models
9.6. Mean Response Models
Chapter Notes
Problems
10. Models for Matched Pairs
10.1. Comparing Dependent Proportions
10.2. Symmetry Models
10.3. Marginal Homogeneity
10.4. Square Tables with Ordered Categories
10.5. Measuring Agreement
10.6. Bradley-Terry Model for Paired Comparisons
Chapter Notes
Problems
11. Analyzing Repeated Categorical Response Data
11.1. Symmetry
11.2. Marginal Homogeneity
11.3. Modeling a Repeated Categorical Response
11.4. Modeling a Repeated Ordinal Response
11.5. Markov Chain Models
Chapter Notes
Problems
12. Asymptotic Theory for Parametric Models
12.1. Delta Method
12.2. Asymptotic Distributions of Estimators of Model Parameters and Cell Probabilities
12.3. Asymptotic Distribution of Residuals and Goodness-of-Fit Statistics
12.4. Asymptotic Distributions for Loglinear Models
Chapter Notes
Problems
13. Estimation Theory for Parametric Models
13.1. Maximum Likelihood for Generalized Linear Models
13.2. Maximum Likelihood for Loglinear Models
13.3. Weighted Least Squares for Categorical Data
13.4. Bayesian Inference for Categorical Data
13.5. Other Methods of Estimation
Chapter Notes
Problems
Appendix A. Using Computer Software to Analyze Categorical Data
A.1. Software Packages
A.2. Listings of Computer Routines by Chapter
Appendix B. A Twentieth-Century Tour of Categorical Data Analysis
Appendix C. Chi-Squared Distribution Values for Various Right-Hand Tail Probabilities
Bibliography
Index of Examples
Index of Selected Notation
Author Index
Subject Index