Contents
Preface ix
Chapter 1 Introduction
to SAS
1.1 What Is SAS Enterprise Guide? 2
1.2 Using This Book 3
1.3 The SAS Enterprise Guide Interface 4
1.3.1 SAS
Enterprise Guide Projects 5
1.3.2
The User Interface 5
1.3.3
The Process Flow 6
1.3.4
The Active Data Set 8
1.4 Creating a Project 9
1.4.1 Opening
a SAS Data Set 9
1.4.2
Importing Data 10
1.5 Modifying Data 15
1.5.1
Modifying Variables: Using Queries
15
1.5.2
Recoding Variables 18
1.5.3
Splitting Data Sets: Using Filters
20
1.5.4
Concatenating and Merging Data Sets:
Appends and Joins 21
1.5.5
Names of Data Sets and Variables in SAS and
SAS Enterprise Guide 26
1.5.6
Storing SAS Data Sets: Libraries
27
1.6 Statistical Analysis Tasks 28
1.7 Graphs 30
1.8 Running Parts of the Process Flow 30
Chapter 2 Data Description and Simple Inference 31
2.1 Introduction 32
2.2 Example: Guessing the Width of a Room: Analysis
of Room
Width Guesses 32
2.2.1
Initial Analysis of Room Width Guesses Using Simple
Summary Statistics and
Graphics 33
2.2.2
Guessing the Width of a Room: Is There Any Difference in
Guesses Made in Feet and in
Meters? 40
2.2.3
Checking the Assumptions Made When Using Student’s
t-Test and Alternatives to the t-Test 47
2.3 Example: Wave Power
and Mooring Methods 49
2.3.1
Initial Analysis of Wave Energy Data Using Box Plots 50
2.3.2 Wave
Power and Mooring Methods: Do Two Mooring
Methods Differ in Bending
Stress?
54
2.3.3
Checking the Assumptions of the Paired t-Tests 56
2.4 Exercises 57
Chapter 3 Dealing with Categorical Data 61
3.1 Introduction 61
3.2 Example: Horse Race
Winners 62
3.2.1
Looking at Horse Race Winners Using Some Simple
Graphics: Bar Charts and Pie
Charts 62
3.2.2
Horse Race Winners: Does Starting Stall Position Predict
Horse Race Winners? 66
3.3 Example: Brain
Tumors 68
3.3.1
Tabulating the Brain Tumor Data into a Contingency
Table 69
3.3.2 Do
Different Types of Brain Tumors Occur More
Frequently at Particular Sites?
The Chi-Square Test 70
3.4 Example: Suicides
and Baiting Behavior 71
3.4.1
How Is Baiting Behavior at Suicides Affected by Season?
Fisher’s Exact Test 71
3.5 Example: Juvenile
Felons 74
3.5.1
Juvenile Felons: Where Should They Be Tried?
McNemar’s Test 75
3.6 Exercises 74
Chapter 4 Dealing with Bivariate Data 79
4.1 Introduction 80
4.2 Example: Heights
and Resting Pulse Rates 80
4.2.1
Plotting Heights and Resting Pulse Rates:
The Scatterplot 81
4.2.2
Quantifying the Relationship between Resting Pulse Rate
and Height: The Correlation
Coefficient 82
4.2.3
Heights and Resting Pulse Rates: Simple Linear
Regression 85
4.3 Example: An
Experiment in Kinesiology 90
4.3.1
Oxygen Uptake and Expired Ventilation:
The Scatterplot 91
4.3.2
Expired Ventilation and Oxygen Uptake: Is Simple Linear
Regression Appropriate? 93
4.4 Example:
4.4.1
Plotting the Birthrate Data: The Aspect Ratio of a
Scatterplot 95
4.5 Exercises 102
Chapter 5 Analysis of Variance 107
5.1 Introduction 108
5.2 Example: Teaching
Arithmetic 108
5.2.1
Initial Examination of the Teaching Arithmetic Data with
Summary Statistics and Box
Plots 109
5.2.2
Teaching Arithmetic: Are Some Teaching Methods for
Teaching Arithmetic Better Than
Others? 112
5.3 Example: Weight
Gain in Rats 116
5.3.1 A
First Look at the Rat Weight Gain Data Using Box Plots
and Numerical Summaries 116
5.3.2 Weight
Gain in Rats: Do Rats Gain More Weight on a
Particular Diet? 119
5.4 Example: Mother’s
Post-Natal Depression and Child’s IQ
124
5.4.1
Summarizing the Post-Natal Depression Data 125
5.4.2
How Is a Child’s IQ Affected by Post-Natal Depression in
the Mother? 128
5.5 Exercises 133
Chapter 6 Multiple Linear Regression 139
6.1 Introduction 140
6.2 Example: Consuming
Ice Cream 140
6.2.1
The Ice Cream Data: An Initial Analysis Using
Scatterplots 141
6.2.2
Ice Cream Sales: Are They Most Affected by Price or
Temperature? How to Tell Using
Multiple Regression 143
6.2.3
Diagnosing the Multiple Regression Model Fitted to the
Ice Cream Consumption Data: The
Use of Residuals 146
6.3 Example: Making It
Rain by Cloud Seeding 152
6.3.1 The Cloud Seeding Data: Initial Examination of
the Data
Using Box Plots and
Scatterplots 154
6.3.2
When Is Cloud Seeding Best Carried Out? How to Tell
Using Multiple Regression
Models Containing Interaction
Terms 158
6.3.3
Diagnosing the Fitted Model for the Cloud Seeding Data
Using Residuals 164
6.4 Exercises 166
Chapter 7 Logistic Regression 171
7.1 Introduction 172
7.2 Example: Myocardial
Infarctions 172
7.2.1
Myocardial Infarctions: What Predicts a Past History of
Myocardial Infarctions?
Answering the Question Using
Logistic Regression 174
7.2.2
Odds 174
7.2.3
Applying the Logistic Regression Model with a Single
Explanatory Variable 175
7.2.4 Interpreting
the Regression Coefficient in the Fitted
Logistic Regression Model 179
7.2.5
Applying the Logistic Regression Model Using
SAS Enterprise Guide 180
7.3 Exercises 186
Chapter 8 Survival Analysis 191
8.1 Introduction 192
8.2 Example: Gastric
Cancer 192
8.2.1
Gastric Cancer Patients: Summarizing and Displaying
Their Survival Experience Using
the Survival
Function 193
8.2.2
Plotting Survival Functions Using SAS Enterprise
Guide 194
8.2.3
Testing the Equality of Two Survival Functions:
The Log-Rank Test 202
8.3 Example:
Myeloblastic Leukemia 204
8.3.1
What Affects Survival in Patients with Leukemia?
The Hazard Function and Cox
Regression 207
8.3.2
Applying Cox Regression Using SAS Enterprise
Guide 209
8.4 Exercises 213
References 215
Index 217