Type II error

Last revised by Henry Knipe on 6 Apr 2015

Type II errors (beta errors, β) occur when we accept that there is no difference between two experimental groups, when in fact, there is a difference.

The p-value does not give a direct indication of the likelihood of a type II error; if the p-value is >0.05, this does not necessarily mean that the two experimental groups are similar. This would only be true if the study is powered to a size to detect a difference if it were present 2.

By convention, we try to keep β = 0.15-0.20

Factors that affect the β:

  • the size of the difference between the groups
  • variability within groups
  • the likelihood of committing a type I error
  • sample size: the larger study, the more power and the less likelihood of a β error

ADVERTISEMENT: Supporters see fewer/no ads

Cases and figures

  • Figure 1: alpha and beta error
    Drag here to reorder.