Required Sample Size (per group)
64
You need 64 subjects per group to have an 80% chance of detecting a medium effect (d=0.5) at α=0.05.
Statistical power is the probability of detecting an effect if it truly exists. A common target is 80%. Use this tool to plan sample sizes before running experiments.
Cohen's d measures the standardized difference between two means. d=0.2 is small, 0.5 medium, 0.8 large.
It balances the risk of Type II error with practical sample size constraints.
It finds the smallest effect size you could detect with your given sample size and power.