Power (stats)

P
 

Statistical power is the likelihood of a hypothesis test of detecting an effect if the effect is present. Statistical power is the probability, or likelihood, that a hypothesis test will correctly reject the null hypothesis. The higher the power the less probable the occurrence of a false negative, or type II error.  Power is impacted by sample size and variance (that is, the actual difference in means between two groups).

See also: hypothesis, null hypothesis, type II error

Return to the Evaluation Dictionary

 
Previous
Previous

Data quality

Next
Next

Pilot