Paired t-test sample size calculator
Enter the standardized difference effect dz, significance level and target power. The result is a planning approximation for complete pairs.
How do I use this page?
- Set dz from pilot data or the minimum meaningful difference.
- Choose alpha, power and direction.
- Add a separate allowance for missing pairs.
What this page does not do: This is a planning approximation for a paired t test. It does not account for dropout, repeated measurements or multiple-testing corrections.
Further reading
Theory — how to calculate sample size for a paired t-test?
A paired t-test analyzes differences within each pair of measurements, for example before and after an intervention. The standardized effect dz is the expected mean difference divided by the standard deviation of those differences, not by the standard deviation of each series separately. The calculator derives the sample size from the normal approximation as a starting point for further refinement. The result represents the number of complete pairs, which in a before–after design usually means the number of people with both measurements.
Frequently asked questions
Does the result mean people or pairs?
Can I use Cohen's d for independent groups?
See also
A/B test calculator
Compare two proportions in an A/B test. Calculate conversion rates, difference, uplift, z score, p value and c
A/B test planning — sample size, MDE and duration
Plan a two-proportion experiment, estimate detectable effect and traffic duration.
ACF and PACF autocorrelation calculator
Calculate ACF and PACF across time-series lags. Review the chart, reference bounds and coefficient table.