Effect size calculator

Calculate Cohen's d and Hedges' g for a difference in means. The effect size shows how much the difference matters in practice, not only whether it is statistically significant.

Effect size — a standardized measure of the difference between groups, expressed in units of the pooled standard deviation.

Calculation FAQ

Group summary data

Control or old version

Standard deviation

Number of observations

Test or new version

Standard deviation

Number of observations

Cohen's d
—
Result
effect size

Interpretation

Change the inputs to recalculate.

In SurveyNinja, NPS, CSAT and more are calculated automatically. Create a survey in 5 minutes and get real-time analytics.

Start free

Data is not stored — calculation runs in your browser.

How Cohen's d is calculated

Formula

Cohen's d = (meanB − meanA) / pooled SD. Hedges' g = d × a small-sample correction.

Cohen's d

Shows by how many standard deviations the group means differ.

Hedges' g

Similar to d, with a correction for small samples.

Practical significance

Even a statistically significant result can be too small to act on.

Calculation examples

1 New onboarding

A mean 7.1, SD 1.4
B mean 7.8, SD 1.3
A d of about 0.5 is a medium effect, already worth discussing with the business.

2 New call-to-action text

A mean 4.2
B mean 4.3
SD about 1.1
A small d can be significant on a large sample yet weak in practice.

What to avoid

  1. Looking only at the p-value The p-value is about statistical confidence; the effect size is about the scale of the difference.
  2. Comparing metrics without context d standardizes the difference, but the business meaning of different metrics still differs.
  3. Ignoring the direction A positive d here means group B is higher than A. Check that this is the direction you want.

Frequently asked questions

Automate your metrics

In SurveyNinja, NPS, CSAT and more are calculated automatically. Create a survey in 5 minutes and get real-time analytics.

Start free