Power & Sample Size for t-Tests

The standardised difference you want to detect: the difference in means divided by the standard deviation.

Rules of thumb: 0.2 small, 0.5 medium, 0.8 large. Smaller effects need much larger samples.

The false-positive rate of the test, usually 0.05. A stricter (smaller) α requires more subjects to reach the same power.

The probability of detecting the effect if it truly exists. The convention is 0.8; higher power needs larger samples.

The sample size of your design — per group for the two-sample test. The tool reports the power this n achieves for the given effect size and α.