StatGardenREF. DESK
Calculators/Statistics/Statistical power analysis
Statistics

Statistical power analysis calculator

Achieved power for a given sample size, and the sample size needed for a target power.

What this calculator does

Statistical power analysis works out achieved power for a given sample size, and the sample size needed for a target power. Enter your own figures above and the answer updates as you type: nothing is fixed in the code, so the result reflects exactly the numbers you supply.

The formula this calculator evaluates is printed under the tool and explained below, so you can check the working by hand or reuse it in a spreadsheet.

The formula

FormulaPower = Φ(d·√(n/2) − z_(α/2)). Required n per group = 2·[(z_(α/2) + z_power) / d]², for a two-sample z-test

The inputs explained

FieldWhat to enter
Effect size (Cohen's d)A number. Starts at 0.5.
Significance level α (two-sided) (%)A number, measured in %. Starts at 5.
Sample size per group (for achieved power)A number. Starts at 40.
Target power (for required n) (%)A number, measured in %. Starts at 80.

Worked examples

Every figure in the tables below is produced by this page’s own calculator at build time, so the numbers and the tool always agree. Select any row to load that scenario.

How the answer changes with effect size (cohen's d)

Every other input is held at the calculator’s starting values while effect size (cohen's d) varies. Select any row to load that scenario into the calculator.

How the answer changes with effect size (cohen's d)
Effect size (Cohen's d)Achieved power at n = 40 per groupSample size needed per group for 80.0% powerCritical z (two-sided α)
0.2520.0%2521.960
0.3839.7%1091.960
0.560.9%631.960
0.7591.8%281.960
199.4%161.960
1.5100.0%71.960