Content: Cover
Half Title
Title Page
Copyright Page
Contents
Preface
Chapter 1: Study Designs and Measures of Effect Size
1.1 Study Designs
1.1.1 Introduction
1.1.2 Nonexperimental or Observational Studies
1.1.3 Types of Nonexperimental Designs
1.1.3.1 Descriptive/Exploratory Survey Studies
1.1.3.2 Correlational Studies (Ecological Studies)
1.1.3.3 Cross-Sectional Studies
1.1.3.4 Longitudinal Studies
1.1.3.5 Prospective or Cohort Studies
1.1.3.6 Case-Control Studies
1.1.3.7 Nested Case-Control Study
1.1.3.8 Case-Crossover Study
1.1.4 Quasi-Experimental Designs 1.1.5 Single-Subject Design (SSD)1.1.6 Quality of Designs
1.1.7 Confounding
1.1.8 Sampling
1.1.9 Types of Sampling Strategies
1.1.10 Summary
1.2 Effect Size
1.2.1 What Is Effect Size?
1.2.2 Why Report Effect Sizes?
1.2.3 Measures of Effect Size
1.2.4 What Is Meant by "Small," " Medium," and " Large" ?
1.2.5 Summary
1.2.6 American Statistical Association (ASA) Statement about the p-value
Exercises
Chapter 2: Comparing Group Means When the Standard Assumptions Are Violated
2.1 Introduction
2.2 Nonnormality
2.3 Heterogeneity of Variances
2.3.1 Bartlett' s Test 2.3.2 Levene' s Test (1960)2.4 Testing Equality of Group Means
2.4.1 Welch' s Statistic (1951)
2.4.2 Brown and Forsythe Statistic (1974b) for Testing Equality of Group Means
2.4.3 Cochran's (1937) Method of Weighing for Testing Equality of Group Means
2.5 Nonindependence
2.6 Nonparametric Tests
2.6.1 Nonparametric Analysis of Milk Data Using SAS
Chapter 3: Analyzing Clustered Data
3.1 Introduction
3.2 The Basic Feature of Cluster Data
3.3 Effect of One Measured Covariate on Estimation of the Intracluster Correlation
3.4 Sampling and Design Issues
3.4.1 Comparison of Means 3.5 Regression Analysis for Clustered Data3.6 Generalized Linear Models
3.6.1 Marginal Models (Population Average Models)
3.6.2 Random Effects Models
3.6.3 Generalized Estimating Equation (GEE)
3.7 Fitting Alternative Models for Clustered Data
3.7.1 Proc Mixed for Clustered Data
3.7.2 Model 1: Unconditional Means Model
3.7.3 Model 2: Including a Family Level Covariate
3.7.4 Model 3: Including the Sib-Level Covariate
3.7.5 Model 4: Including One Family Level Covariate and Two Subject Level Covariates
Appendix
Linear Combinations of Random Variables
Two Linear Combinations The Delta MethodExercises
Chapter 4: Statistical Analysis of Cross-Classified Data
4.1 Introduction
4.2 Measures of Association in 2 ร 2 Tables
4.2.1 Absolute Risk
4.2.2 Risk Difference
4.2.3 Attributable Risk
4.2.4 Relative Risk
4.2.5 Odds Ratio
4.2.6 Relationship between Odds Ratio and Relative Risk
4.2.7 Incidence Rate and Incidence Rate Ratio As a Measure of Effect Size
4.2.8 What Is Person-Time?
4.3 Statistical Analysis from the 2 ร 2 Classification Data
4.3.1 Cross-Sectional Sampling
4.3.2 Cohort and Case-Control Studies
4.4 Statistical Inference on Odds Ratio