The Essential Guide to Effect Sizes: Statistical Power, Meta-Analysis, and the Interpretation of Research Results

Cover......Page 1
Half-title......Page 3
Title......Page 5
Copyright......Page 6
Dedication......Page 7
Contents......Page 9
Figures......Page 11
Tables......Page 12
Boxes......Page 13
Introduction......Page 15
Part I Effect sizes and the interpretation of results......Page 21
Practical versus statistical significance......Page 23
The concept of effect size......Page 24
Two families of effects......Page 26
The d family: assessing the differences between groups......Page 27
The r family: measuring the strength of a relationship......Page 31
Calculating effect sizes......Page 32
1. Specify the effect size index......Page 36
2. Quantify the precision of the estimate using confidence intervals......Page 37
3. Report effects in jargon-free language......Page 41
Summary......Page 44
Notes......Page 45
An age-old debate – rugby versus soccer......Page 51
The problem of interpretation......Page 52
The importance of context......Page 55
The contribution to knowledge......Page 58
Cohen’s controversial criteria......Page 60
Summary......Page 62
Notes......Page 63
Part II The analysis of statistical power......Page 65
The improbable null......Page 67
The need for error insurance......Page 68
Statistical power......Page 72
The analysis of statistical power......Page 76
Prospective power analyses......Page 77
The perils of post hoc power analyses......Page 78
Setting sample sizes......Page 81
Minimum detectable effects......Page 83
Power and precision......Page 84
Summary......Page 86
Notes......Page 87
Surveying the power of the field......Page 93
The curse of multiplicity......Page 98
Statistical power and errors of gullibility......Page 99
How to boost statistical power......Page 101
Summary......Page 102
Notes......Page 103
Part III Meta-analysis......Page 107
The problem of discordant results......Page 109
Reviewing past research – two approaches......Page 110
1. The narrative review – warts and all......Page 111
2. Meta-analysis as a means for generalizing results......Page 114
Meta-analysis in six (relatively) easy steps......Page 117
Meta-analysis and replication research......Page 129
Meta-analysis as a tool for theory development......Page 131
Summary......Page 132
Notes......Page 133
Four ways to ruin a perfectly good meta-analysis......Page 136
Reporting bias and the file drawer problem......Page 137
Publication bias......Page 139
Quantifying the threat of the availability bias......Page 140
2. Include bad results......Page 142
Mixing good and bad studies......Page 143
Mixing good and bad results......Page 144
3. Use inappropriate statistical models......Page 147
The consequences of using the wrong model......Page 149
4. Run analyses with insufficient statistical power......Page 150
Summary......Page 151
Notes......Page 152
Last word: thirty recommendations for researchers......Page 154
Appendix 1 Minimum sample sizes......Page 158
Combining d effects using Hedges et al.\'s method......Page 161
Combining r effects using Hedges et al.\'s method......Page 166
Comparing Hedges et al. with Hunter and Schmidt......Page 169
Notes......Page 172
Bibliography......Page 173
Index......Page 190