Effect Size Calculator
Calculate effect size using your data.
Loading calculator…
What this tool does
Calculate effect size using your data. Cohen's 0.2 / 0.5 / 0.8 benchmarks are rough conventions, not thresholds of importance — a 'small' effect can matter (medicine, conversion rates) and a 'large' one can be trivial.
How to use the Effect Size Calculator
- Enter or select mean — group 1.
- Enter or select mean — group 2.
- Enter or select standard deviation — group 1.
- Enter or select standard deviation — group 2.
- Enter or select sample size — group 1.
- Enter or select sample size — group 2.
- Read the calculated result; change any measurement to compare alternatives.
Formula
pooled SD = sqrt( ((n₁−1)s₁² + (n₂−1)s₂²) / (n₁+n₂−2) ); Cohen's d = (mean₁ − mean₂)/pooled SD; Hedges' g = d × (1 − 3/(4(n₁+n₂−2) − 1))
- m1
- Mean — group 1
- m2
- Mean — group 2
- s1
- Standard deviation — group 1
- s2
- Standard deviation — group 2
- n1
- Sample size — group 1
- n2
- Sample size — group 2
Cohen's 0.2 / 0.5 / 0.8 benchmarks are rough conventions, not thresholds of importance — a 'small' effect can matter (medicine, conversion rates) and a 'large' one can be trivial.
Worked example
For effect size calculator, the following measurements illustrate the exact method: Mean — group 1: 50; Mean — group 2: 45; Standard deviation — group 1: 10; Standard deviation — group 2: 10; Sample size — group 1: 30; Sample size — group 2: 30.
Inputs
- Mean — group 150
- Mean — group 245
- Standard deviation — group 110
- Standard deviation — group 210
- Sample size — group 130
- Sample size — group 230
Result
- Cohen's d0.5
- Hedges' g (bias-corrected)0.49
- Pooled standard deviation10
- Conventional size labelMedium
Results explained
- Cohen's d
- Cohen's d from the formula above. Cohen's 0.2 / 0.5 / 0.8 benchmarks are rough conventions, not thresholds of importance — a 'small' effect can matter (medicine, conversion rates) and a 'large' one can be trivial.
- Hedges' g (bias-corrected)
- Hedges' g (bias-corrected) from the formula above. Cohen's 0.2 / 0.5 / 0.8 benchmarks are rough conventions, not thresholds of importance — a 'small' effect can matter (medicine, conversion rates) and a 'large' one can be trivial.
- Pooled standard deviation
- Pooled standard deviation from the formula above. Cohen's 0.2 / 0.5 / 0.8 benchmarks are rough conventions, not thresholds of importance — a 'small' effect can matter (medicine, conversion rates) and a 'large' one can be trivial.
- Conventional size label
- Conventional size label from the formula above. Cohen's 0.2 / 0.5 / 0.8 benchmarks are rough conventions, not thresholds of importance — a 'small' effect can matter (medicine, conversion rates) and a 'large' one can be trivial.
Frequently asked questions
pooled SD = sqrt( ((n₁−1)s₁² + (n₂−1)s₂²) / (n₁+n₂−2) ); Cohen's d = (mean₁ − mean₂)/pooled SD; Hedges' g = d × (1 − 3/(4(n₁+n₂−2) − 1))
Cohen's 0.2 / 0.5 / 0.8 benchmarks are rough conventions, not thresholds of importance — a 'small' effect can matter (medicine, conversion rates) and a 'large' one can be trivial.
Enter numbers only, in the units and format each label describes. Remove missing values rather than substituting zero, unless zero is a real observation.
No. Results describe the numbers you entered. Statistical inference also depends on sampling design, independence, model fit and interpretation; a p-value is not the probability that a hypothesis is true.
No. Every calculation, including the distribution algorithms, runs entirely in your browser.