Science and education
StatPower Pro
Power and sample size
A statistical power and sample-size toolkit — 22 calculators covering t-tests, ANOVA, chi-square, regression, equivalence and Bayesian designs.
What StatPower Pro does
StatPower Pro works out how large a study needs to be to detect an effect, and how much power a given sample has. It has 22 calculators in six families:
- Power analysis (9): t-test, paired t-test, correlation, regression, ANOVA, chi-square, two proportions, equivalence (TOST) and reliability
- Applied (4): A/B test, survey sample size, sample-size tables and minimum detectable effect
- Effect sizes (3): lookup, converter and interpreter
- Visualization (4): error simulator, confidence-interval width explorer, one- versus two-sided tests, and p-value distributions
- Bayesian (1): Bayesian sample size
- Reference (1): a comparison of tests
Results can be copied, exported to PDF and cited in APA format. It also has an onboarding tour, saved analyses and a dark mode.
Who it’s for
Researchers, data scientists and product teams planning a study or an experiment. The calculators work without an account; signing in saves analyses.
How the numbers are computed
Every value is computed live in your browser by the same functions that power each calculator; no results are hard-coded. A G*Power Validation page runs a set of documented test cases against G*Power 3.1 and reports each one as an exact match, an approximate match or a mismatch.
What it isn’t
A drop-in replacement for G*Power on every design. The two don’t agree in every case, and the validation page lists the ones where they differ, so check it before relying on a figure for a protocol or a grant application.
