A p-value tells you whether a difference is probably real; the effect size tells you how big it is. Enter two groups' summary stats to get Cohen's d — the difference in standard-deviation units.

How to use it

Enter the mean, standard deviation, and sample size for each of the two groups. The calculator pools the two standard deviations and divides the mean difference by that pool.

The formula. Cohen's d = (x̄₁ − x̄₂) / s_pooled, where the pooled standard deviation is s_pooled = √( ((n₁−1)s₁² + (n₂−1)s₂²) / (n₁+n₂−2) ). The sign just reflects which group is larger; what matters is the magnitude.

Reading the result

\|d\|	Conventional label	Plain meaning
≈ 0.2	Small	Groups overlap a lot; the difference is subtle.
≈ 0.5	Medium	A difference you could notice with the naked eye.
≈ 0.8+	Large	Groups are clearly separated.

These labels are conventions, not laws — in a high-stakes field a "small" effect can matter enormously, and in a noisy one a "large" effect may be unremarkable. Always read d in the context of your domain.

Why this matters. With a large enough sample, even a microscopic difference earns p < 0.05. Effect size is independent of sample size, so reporting d alongside the p-value is what separates "statistically significant" from "actually meaningful." Then feed d into the sample-size calculator to plan your next study.

Effect size calculator (Cohen's d)

How to use it

Reading the result

See also