Skip to main content

Null hypothesis

Null hypothesis is the default claim in a hypothesis test, asserting no effect or a specific parameter value. It is treated as true and then challenged with sample evidence.

Also known asH0

ByHoang TruongUpdated

FrameworkHypothesis testing

See it move

Loading infographic...

The infographic is a formula card presenting two standard null-hypothesis forms: H₀: μ = μ₀ when testing a population mean, and H₀: β₁ = 0 when testing whether a regression slope is zero. Three annotation points follow: the null is assumed true from the start, it is never 'accepted' but only 'failed to reject' when evidence is weak, and it is rejected when the p-value falls at or below the chosen significance level α.

Where it fits
SubjectData Analysis & StatisticsCoreTopicHypothesis TestingCore

Check yourself

PracticeCORE

A bank tests whether its new digital interface changes the average transaction time compared with the previous system's mean of 45 seconds. A sample of 60 transactions gives a p-value of 0.21. Which conclusion is correct at the 5% significance level?

Select an answer to check your understanding.
Null Hypothesis — Statistical Testing