(1) An assumption or proposition where an observed difference between two samples of a statistical population is purely accidental and not due to systematic causes.
(2) The hypothesis to be investigated through statistical hypothesis testing so that when refuted indicates that the alternative hypothesis is true.
A null hypothesis is a hypothesis that is valid or presumed true until invalidated by a statistical test. Thus, one cannot say that a null hypothesis is ‘’accepted’’ rather a null hypothesis ‘’cannot be rejected’’ as it remains to be true based on the statistical evidence supporting it. Otherwise, a null hypothesis that is refuted is said to be ‘’rejected’’.
Word origin: coined by the English geneticist and statistician Ronald Fisher.
Compare: alternative hypothesis
See also: statistics, hypothesis