Sample Size Calculation - Means

one-sided test

two-sided test

ratio of: (effect to detect) / (standard deviation of process)

δ/σ

0.5

(α,β)

α: risk of rejecting H 0 when true	0.05
1-α: confidence	0.95
β: risk of accepting H 0 when false	0.2
1-β: power of the test	0.8
n: sample size	33

This Demonstration calculates the minimum sample size,

n

, required to be randomly drawn from a population, to be able to detect a difference in the population mean of at least

δ

units from the true population mean with confidence of

1-α

. You can vary

α

(the Type I risk),

β

(the Type II risk), or

δ/σ

, the ratio of the detected difference to the population standard deviation. In most practical cases, the standard deviation may not actually be known, but is estimated from prior knowledge or experience.