Categorical Data Analysis Using the SAS System

by Stokes, Maura E.; Davis, Charles S.; Koch, Gary G.

• ISBN13: 9780471224242
• ISBN10: 0471224243
• Edition: 2nd
• Format: Paperback
• Copyright: 2001-12-26
• Publisher: Wiley-SAS

Summary

Along with providing a useful discussion of categorical data analysis techniques, this book shows how to apply these methods with the SAS System. The authors include practical examples from a broad range of applications to illustrate the use of the FREQ, LOGISTIC, GENMOD, and CATMOD procedures in a variety of analyses. They also discuss other procedures such as PHREG and NPAR1WAY.

Author Biography

Maura E. Stokes is Senior Manager of Statistical Applications Research and Development at SAS Institute. She received her PhD in Biostatistics from the University of North Carolina at Chapel Hill and has taught and written about categorical data analysis for over fifteen years.<br> <br> Charles S. Davis is Professor of Biostatistics at the University of Iowa. He received his PhD in Biostatistics from the University of Michigan. His research and teaching interests include categorical data analysis and methods for the analysis of repeated measures.<br> <br> och is Professor of Biostatistics and Director of the biometrics Consulting Laboratory at the University of North Carolina at Chapel Hill. He has had a prominent roll in the field of categorical data analysis for the last thirty years. He teaches classes and seminars in categorical data analysis, consults in areas of statistical practice, conducts research, and trains many Biostatistics students.

Table of Contents

Preface to the Second Edition	v
Acknowledgments	vii
Introduction	1	(16)
Overview	3	(1)
Scale of Measurement	3	(3)
Sampling Frameworks	6	(1)
Overview of Analysis Strategies	7	(3)
Working with Tables in the SAS System	10	(5)
Using This Book	15	(2)
The 2 x 2 Table	17	(26)
Introduction	19	(1)
Chi-Square Statistics	20	(3)
Exact Tests	23	(6)
Difference in Proportions	29	(3)
Odds Ratio and Relative Risk	32	(7)
Sensitivity and Specificity	39	(1)
McNemar's Test	40	(3)
Sets of 2 x 2 Tables	43	(22)
Introduction	45	(1)
Mantel-Haenszel Test	45	(12)
Measures of Association	57	(8)
Sets of 2 x r and s x 2 Tables	65	(24)
Introduction	67	(1)
Sets of 2 x r Tables	67	(11)
Sets of s x 2 Tables	78	(8)
Relationships Between Sets of Tables	86	(3)
The s x r Table	89	(32)
Introduction	91	(1)
Association	91	(9)
Exact Tests for Association	100	(5)
Measures of Association	105	(6)
Observer Agreement	111	(5)
Test for Ordered Differences	116	(5)
Sets of s x r Tables	121	(38)
Introduction	123	(1)
General Mantel-Haenszel Methodology	124	(3)
Mantel-Haenszel Applications	127	(10)
Advanced Topic: Application to Repeated Measures	137	(22)
Nonparametric Methods	159	(22)
Introduction	161	(1)
Wilcoxon-Mann-Whitney Test	161	(4)
Kruskal-Wallis Test	165	(3)
Friedman's Chi-Square Test	168	(2)
Aligned Ranks Test for Randomized Complete Blocks	170	(1)
Durbin's Test for Balanced Incomplete Blocks	171	(3)
Rank Analysis of Covariance	174	(7)
Logistic Regression I: Dichotomous Response	181	(60)
Introduction	183	(1)
Dichotomous Explanatory Variables	184	(11)
Using the CLASS Statement	195	(8)
Qualitative Explanatory Variables	203	(8)
Continuous and Ordinal Explanatory Variables	211	(6)
A Note on Diagnostics	217	(5)
Maximum Likelihood Estimation Problems and Alternatives	222	(3)
Exact Methods in Logistic Regression	225	(7)
Using the CATMOD and GENMOD Procedures for Logistic Regression	232	(9)
Appendix A: Statistical Methodology for Dichotomous Logistic Regression	239	(2)
Logistic Regression II: Polytomous Response	241	(30)
Introduction	243	(1)
Ordinal Response: Proportional Odds Model	243	(14)
Nominal Response: Generalized Logits Model	257	(14)
Conditional Logistic Regression	271	(52)
Introduction	273	(1)
Paired Observations from a Highly Stratified Cohort Study	273	(3)
Clinical Trials Study Analysis	276	(7)
Crossover Design Studies	283	(12)
General Conditional Logistic Regression	295	(5)
Paired Observations in a Retrospective Matched Study	300	(9)
1:m Conditional Logistic Regression	309	(5)
Exact Conditional Logistic Regression in the Stratified Setting	314	(9)
Appendix A: Theory for the Case-Control Retrospective Setting	318	(2)
Appendix B: Theory for Exact Conditional Inference	320	(1)
Appendix C: ODS Macro	321	(2)
Quantal Bioassay Analysis	323	(24)
Introduction	325	(1)
Estimating Tolerance Distributions	325	(5)
Comparing Two Drugs	330	(9)
Analysis of Pain Study	339	(8)
Poisson Regression	347	(16)
Introduction	349	(1)
Methodology for Poisson Regression	349	(2)
Simple Poisson Counts Example	351	(2)
Poisson Regression for Incidence Densities	353	(3)
Overdispersion in Lower Respiratory Infection Example	356	(7)
Weighted Least Squares	363	(64)
Introduction	365	(1)
Weighted Least Squares Methodology	365	(6)
Using PROC CATMOD for Weighted Least Squares Analysis	371	(6)
Analysis of Means: Performing Contrast Tests	377	(9)
Analysis of Proportions: Occupational Data	386	(9)
Obstetrical Pain Data: Advanced Modeling of Means	395	(14)
Analysis of Survey Sample Data	409	(9)
Modeling Rank Measures of Association Statistics	418	(9)
Appendix A: Statistical Methodology for Weighted Least Squares	422	(5)
Modeling Repeated Measurements Data with WLS	427	(42)
Introduction	429	(1)
Weighted Least Squares	430	(23)
Advanced Topic: Further Weighted Least Squares Applications	453	(16)
Generalized Estimating Equations	469	(82)
Introduction	471	(1)
Methodology	471	(7)
Summary of the GEE Methodology	478	(2)
Passive Smoking Example	480	(7)
Crossover Example	487	(7)
Respiratory Data	494	(9)
Using a Modified Wald Statistic to Assess Model Effects	503	(2)
Diagnostic Data	505	(5)
Using GEE for Count Data	510	(4)
Fitting the Proportional Odds Model	514	(4)
GEE Analyses for Data with Missing Values	518	(9)
Alternating Logistic Regression	527	(6)
Using GEE to Fit a Partial Proportional Odds Model: Univariate Outcome	533	(8)
Using GEE to Account for Overdispersion: Univariate Outcome	541	(10)
Appendix A: Steps to Find the GEE Solution	547	(1)
Appendix B: Macro for Adjusted Wald Statistic	548	(3)
Loglinear Models	551	(40)
Introduction	553	(1)
Two-Way Contingency Tables	554	(10)
Three-Way Contingency Tables	564	(10)
Higher-Order Contingency Tables	574	(11)
Correspondence Between Logistic Models and Loglinear Models	585	(6)
Appendix A: Equivalence of the Loglinear and Poisson Regression Models	588	(3)
Categorized Time-to-Event Data	591	(16)
Introduction	593	(1)
Life Table Estimation of Survival Rates	593	(3)
Mantel-Cox Test	596	(3)
Piecewise Exponential Models	599	(8)
References	607	(12)
Index	619