type 2 error

Last edited 06/2020

A type 2 error is an error resulting from wrongly accepting Ho - the null hypothesis - when it is in fact false (i.e. a false negative result).

  • the power of a study can be calculated via knowledge of the type 2 error (1)

    • power (the chance that the study will detect the minimum difference as statistically different) = 1 - type 2 error

The power of a study of a clinical trial (1 - probability of a type 2 error) simultaneously reflects the number of people in the trial and the amount of actual change in a particular subgroup.

  • rate of the type II error is denoted by the Greek letter ß (beta) and related to the power of a test (which equals 1-ß)

Reference:

  1. Evidence Based Medicine 2006;11:69-70.