Update: We are now collecting Disclosures from all users who edit content. You will be prompted to update these when you next edit content or you can complete your disclosures at any time in your user profile. This has been reflected in an updated terms-of-use.

Type I error

Last revised by Dr Daniel J Bell on 14 Nov 2017

Type I errors (alpha errors, α) occur when we accept that there is a difference between two experimental groups, when in fact, no difference exists.

The threshold for accepting a type I error is the p-value. The traditionally accepted p-value of 0.05 indicates that the researchers are willing to take a ≤5% chance that the outcome of their data showing a difference between two groups was merely due to chance.

If multiple tests are compared, the risk of type I error increases.

Good statistical technique attempts to keep the risk of type I error as low as possible.

ADVERTISEMENT: Supporters see fewer/no ads

Cases and figures

  • Figure 1: Alpha error
    Drag here to reorder.
  • Figure 2: alpha and beta error
    Drag here to reorder.
  • Updating… Please wait.

     Unable to process the form. Check for errors and try again.

     Thank you for updating your details.