A rejection region is the set of values of a test statistic for which a statistical test rejects its null
hypothesis. It is chosen so that, when the null hypothesis is true, the probability
that the test statistic falls in the rejection region is at most the significance
level 
.
For a continuous test statistic 
in a right-tailed test, a rejection region often has the form
, where the critical
value 
satisfies
 |
For a discrete test statistic, equality may be impossible, so the actual rejection probability can be smaller than 
. The rejection region is the complement, up to boundary
conventions, of the acceptance region.
