Cohen’s d Effect Size: How to Quantify the Size of a Difference

Why Effect Size Matters

A p-value tells you whether your data are compatible with a null hypothesis, but it does not tell you how large the observed difference is. That is where effect size becomes useful. Cohen’s d is one of the most common effect size measures for comparing two group means.

It expresses the mean difference in standard deviation units, which makes the result easier to compare across studies or outcomes measured on different scales.

The Formula

A common version of Cohen’s d is:

d = (Mean1 - Mean2) / pooled standard deviation

If Group 1 is one standard deviation higher than Group 2, then d = 1.0. If the groups differ by half of a standard deviation, then d = 0.5.

When Should You Use Cohen’s d?

• When comparing two means.
• When you want to communicate practical magnitude, not just significance.
• When readers may benefit from a standardized summary.

Step-by-Step Method

1. Calculate the mean for each group.
2. Compute the standard deviation for each group.
3. Find the pooled standard deviation if the standard two-group version is appropriate.
4. Subtract one mean from the other.
5. Divide that mean difference by the pooled standard deviation.

An Original Example

Suppose a teaching team compares final improvement scores between a workshop group and a self-study group. The workshop group has a mean improvement of 16 points, while the self-study group has a mean improvement of 11 points. The pooled standard deviation is 6.7.

The effect size is:

d = (16 - 11) / 6.7 = 0.75

An effect size of 0.75 suggests a moderately large difference. The workshop group improved by about three quarters of a standard deviation compared with the self-study group.

How to Interpret Cohen’s d

Introductory textbooks often use rough benchmarks such as:

• 0.2: small
• 0.5: medium
• 0.8: large

These labels are convenient, but they should not replace subject-matter judgment. In some settings, an effect size of 0.2 may be meaningful. In other settings, even 0.8 may not justify changing practice if the intervention is expensive or difficult to implement.

Important Practical Points

• Cohen’s d complements hypothesis testing rather than replacing it.
• The sign of d depends on the order of subtraction.
• For small samples or special study designs, adjusted effect size formulas may be preferable.

Common Mistakes

• Reporting significance without any effect size.
• Using the effect size label alone without context.
• Forgetting to define which group was subtracted from which.
• Interpreting d as a percentage difference.

Short Summary

Cohen’s d is a simple and powerful way to show how big a mean difference really is. Whenever you compare two groups, adding this effect size can make your results more useful and easier to understand.