Effect Size Calculator

Measure a standardized effect size with Cohen's d.

Enter at least two values (or two pairs).

How Cohen's d works

Cohen's d is a standardized measure of effect size: the difference between two means (or a mean and a target) divided by a standard deviation. Because it is unit-free, effect sizes from different studies can be compared.

For independent groups, d uses the pooled standard deviation. For a paired design it uses the standard deviation of the differences (sometimes called d_z). For one sample it compares the mean to a target value.

Labels like 'small' (0.2), 'medium' (0.5) and 'large' (0.8) are Cohen's conventions. They are heuristics, not universal scientific rules — what counts as a meaningful effect depends entirely on your field and context.

The formulas

d = (x̄₁ − x̄₂) / s_pooled · s_pooled = √(((n₁−1)s₁² + (n₂−1)s₂²) / (n₁+n₂−2))

For one-sample, d = (x̄ − μ₀)/s. For paired data, d = d̄/s_d. The sign indicates the direction of the effect; the magnitude matters for interpretation.

Worked example

Groups 1,2,3 vs 4,5,6

The pooled standard deviation is 1 and the mean difference is −3, so d = −3. The magnitude 3 is far beyond the 'large' convention (0.8), indicating a very large effect.

Frequently asked questions

Are the 'small, medium, large' labels universal?

No. They are Cohen's widely used conventions and depend on context. A 'small' effect in one field may be important in another — always interpret effect size within your domain.

What does the sign of d mean?

It indicates direction: a positive d means the first group (or sample) has the higher mean. For interpretation, the absolute value is what matters.

Why report effect size alongside a p-value?

A p-value tells you whether an effect is likely real; an effect size tells you how large it is. A statistically significant result can still have a trivially small effect.

Built by PanelRoster

PanelRoster is an enterprise operations management platform for organizations coordinating distributed teams, assignments, workflows, quality assurance and operational execution.

Explore PanelRoster