Cohen’s d Calculator

Work out Cohen’s d, the effect size of the difference between two group means, and whether it is small, medium or large.

Your details

Your data stays in your browser

Result

Enter each group’s mean, SD and size, then press “Calculate”.

Cohen’s benchmarks

Your result will be marked here once calculated.

  1. NegligibleUnder 0.2
  2. Small0.2–0.49
  3. Medium0.5–0.79
  4. Large0.8 or more

Cohen’s benchmarks are guides; what matters in practice depends on the field.

Analysis and suggestions

Once calculated, this shows what your result means and suggestions based on your numbers.

Behind the result

Work out Cohen’s d, the effect size of the difference between two group means, and whether it is small, medium or large.

Example · d = (المتوسط الأول − المتوسط الثاني) ÷ الانحراف المجمّع | الانحراف المجمّع = √(((n₁ − 1)s₁² + (n₂ − 1)s₂²) ÷ (n₁ + n₂ − 2))

d = 6 ÷ 11.0454 = 0.543

How it is calculated

Comprehensive Calculator Guide

This calculator gives Cohen’s d, the difference between two group means in standard-deviation units, showing whether it is small, medium or large regardless of sample size. A new teaching method averaging 78 against 72, with SDs of 10 and 12, has d = 0.54, a medium effect.

Effect size

d = (mean₁ − mean₂) ÷ pooled SD. Cohen’s guides: 0.2 small, 0.5 medium, 0.8 large, but practical importance depends on the field. Significance says a difference is real; effect size says it matters.

How to Use

  1. Enter group 1’s mean, SD and size.
  2. Enter group 2’s.
  3. Press “Calculate” to see d and its size.

Formula Used

d = (x̄₁ − x̄₂) ÷ s_pooled

Practical Examples

Example 1
Teaching: 0.54.
Example 2
Exercise: 1.45.

Important Tips

  • Report d with p and a CI.
  • Interpret in context.
  • Use Hedges’ g for tiny samples.

Common Mistakes to Avoid

  • Using one group’s SD.
  • Dismissing small effects.
  • Confusing d with significance.

Frequently Asked Questions

What is d?
A standardised mean difference.
Large?
0.8+.
Negative?
Yes; read its size.

Read Also

Results are estimates for educational purposes and may vary depending on your situation and data sources.