Statistics & Probability

Instant, private, and free

Cohen's d Calculator.

Calculate a pooled-standard-deviation standardized mean difference.

On-device calculationNo signup
01

Set your values

Results update as you type.

Pooled standard deviation: 11.009736

Pooled standard deviation

11.009736
Cohen's d: 0.454143

Cohen's d

0.454143

Results update automatically as you type.

Use Cases

Compare group means in research studies

Researchers can quantify the magnitude of difference between two experimental or control groups using means, SDs, and sample sizes.

Example: Compare test scores of two teaching methods with n=30 each.

Meta-analysis effect size calculation

When combining results from multiple studies, Cohen's d provides a standardized effect size that can be averaged across studies.

Example: Calculate d for each study to pool in a meta-analysis.

Frequently Asked Questions

What is Cohen's d and how is it calculated?
Cohen's d measures the standardized difference between two group means. It is calculated by subtracting Mean 2 from Mean 1 and dividing by the pooled standard deviation, which is derived from the SDs and sample sizes of both groups.
What does a Cohen's d value indicate?
Cohen's d values are commonly interpreted as: 0.2 = small effect, 0.5 = medium effect, and 0.8 = large effect. However, context matters, and these thresholds are general guidelines.
Can I use this calculator with unequal sample sizes?
Yes, the calculator accepts different sample sizes for each group. The pooled standard deviation formula accounts for unequal sample sizes by weighting each group's SD by its sample size minus one.

Tips & Common Mistakes

Tips

  • Ensure the means and SDs are in the same units for both groups.
  • Use sample standard deviations (n-1) rather than population SDs for accurate pooled estimates.
  • Double-check that sample sizes are entered as whole numbers and match your data.
  • Interpret the result in the context of your field, not just by generic thresholds.

Common Mistakes to Avoid

  • Entering variances instead of standard deviations in the SD fields.
  • Using population SDs (divided by n) instead of sample SDs (divided by n-1).
  • Confusing the order of groups, which changes the sign of d but not its magnitude.

Last updated: August 13, 2026