Help and contact information for BioStat Lab.
Email: datuz2010@gmail.com
Enter the expected mean difference (Δ), standard deviation (σ), significance level (α, typically 0.05), and desired power (1−β, typically 0.80). Select Independent for two separate groups or Paired for matched-pair designs. The result shows the required sample size per group (independent) or number of pairs (paired).
Use the stepper to set the number of groups (2–20). Enter the maximum expected mean difference between any two groups, the within-group standard deviation, α, and power. The result shows both per-group and total sample size.
Enter the two expected proportions (e.g. 0.30 and 0.50 for 30% vs 50%), α, and power. The app shows the required sample size per group along with the absolute difference and risk ratio between the two proportions.
Enter the smallest correlation coefficient (ρ) you want to be able to detect (e.g. 0.30 for a medium effect), α, and power. The calculation uses Fisher's z-transformation. Cohen's effect-size classification (small/medium/large) is shown alongside the result.
Enter the expected hazard ratio (HR), the anticipated proportion of subjects who will experience the event (event rate, 0–1), α, and power. The app uses Freedman's formula to output the minimum number of events required and the total number of subjects needed given the event rate.