A “type one” statistical error occurs when a purely random fluctuation is taken as evidence for a positive effect. The risk of making a “false positive” error of this sort is symbolized by the Greek letter alpha (α). A type 2 statistical error results when we fail to detect an effect when there is one. The risk of making a “false negative” error of this sort is symbolized by the Greek letter beta (β). Often a false negative error may be extremely costly, and so α is set at a very low value, but this increases the risk of type 2 errors. See confidence limits, null hypothesis method, significance of results.
Subjects: Genetics and Genomics.