Power and Sample Size
Definition and Characteristics
In statistical hypothesis testing, the power of the test refers to (1−β), where β is the probability of not rejecting the null hypothesis when it is false. Another component of the statistical comparison is the type I error α which gives the probability of rejecting the null hypothesis when it is true. The preferable analysis method is the one with small error and large power. The sample size and the power of the analysis are closely related and the larger the sample size, the more powerful the testing procedure is.
To design the study, one has to formulate the hypothesis, determine the appropriate testing procedure, set the desired error rate and the power, and estimate the appropriate sample size. In certain types of observational studies or when analyzing previously collected data, the calculations are reversed and power of the test is determined based on the data size.
- Van Belle G (2008) Chapter 2. In: Statistical rules of thumb, 2nd edn. Wiley, Hoboken, NJGoogle Scholar