Table of Contents
What Is Cohen's d?
Cohen's d is the most widely used effect size measure for comparing two group means. It expresses the difference in terms of standard deviation units, making it independent of sample size and measurement scale. Introduced by Jacob Cohen in 1962, it is essential for power analysis and meta-analysis.
Unlike p-values, which only indicate statistical significance, effect sizes quantify the magnitude of the difference, answering "how big is the effect?" rather than just "is there an effect?"
Formula
Cohen's Conventions
| |d| | Effect Size |
|---|---|
| < 0.2 | Negligible |
| 0.2-0.5 | Small |
| 0.5-0.8 | Medium |
| > 0.8 | Large |
FAQ
Why report effect size alongside p-values?
P-values are influenced by sample size. Large samples can produce significant p-values for trivially small effects. Cohen's d gives the practical significance regardless of sample size.
What is the relationship between d and overlap?
Larger d means less overlap between distributions. d=0 means complete overlap (identical groups). d=2 means about 81% of one group exceeds the other group's mean.