In statistical inference, a set of restrictions being tested. The null hypothesis is assumed to be true unless the test suggests otherwise, in which case it is rejected in favour of the alternative hypothesis. The test is usually set such that the distribution of the test statistic, under the assumption that the null hypothesis is true, is known (the null distribution). The null hypothesis is rejected in favour of the alternative if the numerical value of the test statistic falls in the region of values that can be generated by a random draw from the null distribution only with a low probability. See also one-tailed test; p-value; two-tailed test; type I and II errors.