null hypothesis

[noun]

In statistical testing, a null hypothesis (H₀) commonly takes the form of a statement suggesting that an observed result or effect is due to random chance. The null hypothesis is put forward as the counterpart to an alternative hypothesis (H₁), and is presumed correct until statistical testing shows otherwise. For example, in an evaluation of data regarding the pain relieving properties of a new drug, the null hypothesis would state that the new drug has no effect beyond that of a control. The null hypothesis is never proven by statistical testing; either H₁ is accepted in favor of H₀, or H₀ is "not rejected."