Statistical Power Analysis
A Simple and General Model for Traditional and Modern Hypothesis Tests......Page 4
Copyright......Page 5
Contents......Page 6
Preface......Page 10
1 The Power of Statistical Tests......Page 14
The Structure of Statistical Tests......Page 15
The Mechanics of Power Analysis......Page 22
Statistical Power of Research in the Social and Behavioral Sciences......Page 30
Using Power Analysis......Page 32
Hypothesis Tests Versus Confidence Intervals......Page 36
Summary......Page 37
2 A Simple and General Model for Power Analysis......Page 38
The General Linear Model, the F Statistic, and Effect Size......Page 40
The F Distribution and Power......Page 42
Using the Noncentral F Distribution to Assess Power......Page 45
Translating Common Statistics and ES Measures Into F......Page 46
Defining Large, Medium, and Small Effects......Page 51
Nonparametric and Robust Statistics......Page 52
From F to Power Analysis......Page 53
Analytic and Tabular Methods of Power Analysis......Page 54
Using the One-Stop F Table......Page 55
The One-Stop F Calculator......Page 58
Summary......Page 60
3 Power Analyses for Minimum-Effect Tests......Page 62
Implications of Believing That the Nil Hypothesis Is Almost Always Wrong......Page 66
Minimum-Effect Tests as Alternatives to Traditional Null Hypothesis Tests......Page 69
Testing the Hypothesis That Treatment Effects Are Negligible......Page 72
Using the One-Stop Tables to Assess Power to Test Minimum-Effect Hypotheses......Page 77
Using the One-Stop F Calculator for Minimum-Effect Tests......Page 80
Summary......Page 81
4 Using Power Analyses......Page 84
Estimating the Effect Size......Page 85
Four Applications of Statistical Power Analysis......Page 90
Calculating Power......Page 91
Determining Sample Sizes......Page 92
Determining the Sensitivity of Studies......Page 94
Determining Appropriate Decision Criteria......Page 95
Summary......Page 100
5 Correlation and Regression......Page 102
The Perils of Working With Large Samples......Page 103
Multiple Regression......Page 105
Power in Testing for Moderators......Page 109
Why Are Most Moderator Effects Small?......Page 110
Implications of Low Power in Tests for Moderators......Page 112
Summary......Page 113
The t-Test......Page 114
Independent Groups t-Test......Page 116
Traditional Versus Minimum-Effect Tests......Page 118
One-Tailed Versus Two-Tailed Tests......Page 120
Repeated Measures or Dependent t-Test......Page 121
The Analysis of Variance......Page 123
Which Means Differ?......Page 126
Summary......Page 129
7 Multifactor ANOVA Designs......Page 130
The Factorial Analysis of Variance......Page 131
Factorial ANOVA Example......Page 137
Fixed, Mixed, and Random Models......Page 139
Randomized Block ANOVA: An Introduction to Repeated-Measures Designs......Page 141
Independent Groups Versus Repeated Measures......Page 142
Complexities in Estimating Power in Repeated-Measures Designs......Page 147
Summary......Page 148
Split-Plot Factorial ANOVA......Page 150
Power for Within-Subject Versus Between-Subject Factors......Page 153
The Multivariate Analysis of Variance......Page 154
Summary......Page 157
9 The Implications of Power Analyses......Page 158
Tests of the Traditional Null Hypothesis......Page 159
Tests of Minimum-Effect Hypotheses......Page 160
Direct Benefits of Power Analysis......Page 164
Indirect Benefits of Power Analysis......Page 166
Costs Associated With Power Analysis......Page 167
Implications of Power Analysis: Can Power Be Too High?......Page 168
Summary......Page 170
References......Page 172
Appendices......Page 176
Author Index......Page 222
Subject Index......Page 224