Power is the probability of test to correctly reject a false null hypothesis. The ability to detect a significant difference when the do exist.

Power is a function of four things:
1) Level of significance
2) Variance
3) Sample size
4) Effect size

Power is equal to 1-Beta. Beta is the probability of a type 2 error.
Larger studies have more power.

Two purposes of power analysis:
1) Estimate the sample size prior to data collection
2) Calculate the probability that a type 2 error was committed

Comments are closed.