Null hypothesis
Default hypothesis tested in statistical significance tests.
The null hypothesis is a fundamental concept in statistical hypothesis testing, serving as the default statement under scrutiny. Denoted as H₀, it is typically a statement of "no effect" or "no difference," such as the claim that a drug produces zero mean change in blood pressure. The test of significance is designed to assess the strength of the evidence against this null hypothesis. The null hypothesis and its negation, the alternative hypothesis (H₁), are mutually exclusive statements. The research hypothesis is usually consistent with the alternative, though it can sometimes align with the null. In a test, a test statistic is calculated from sample data. The tester then computes the conditional probability of observing a value at least as extreme as the test statistic, assuming the null hypothesis is true. This probability is called the p-value. If the p-value is less than the chosen significance level (α), the null hypothesis is rejected; otherwise, it is not rejected. The test does not conclude that the null hypothesis is false, nor that the probability of its falsehood is less than α. Because sample data cannot definitively prove a hypothesis, conclusions are uncertain. Two types of error are possible: a Type I error, where the null hypothesis is rejected when it is actually true (with probability α, the significance level); and a Type II error, where the null hypothesis is accepted when the alternative is true (with probability β). The quantity 1−β is the power of the test. A hypothesis that completely specifies the population distribution is a simple hypothesis; one that does not is a composite hypothesis. The critical region is the set of test statistic values leading to rejection of the null hypothesis, while the acceptance region is the set for which the null is not rejected. Critical values are the boundaries of the acceptance region. The size of a test is its probability of incorrectly rejecting the null hypothesis; for composite hypotheses, this is the supremum of that probability over all cases covered by the null. A test is conservative if the true probability of a Type I error never exceeds the nominal significance level. The concept of statistical significance predates the formal hypothesis test, originally serving as a pragmatic heuristic for identifying meaningful results. The modern hypothesis test added mathematical rigor by making the
- field
- Statistical inference
- known_for
- Default hypothesis in hypothesis testing, representing no effect or no difference
- type
- Statistical concept
Lore & Background
The null hypothesis is a statement about a population parameter that is tested for possible rejection under a statistical hypothesis test. Typically, it is a default position of "no effect" or "no difference," such as the claim that the mean change in blood pressure from a drug is zero. It is mutually exclusive with the alternative hypothesis, which is its negation. The test is designed to assess the strength of the evidence against the null hypothesis, not to prove it true. A test statistic is calculated from sample data, and the conditional probability of observing a value at least as extreme as the one obtained is computed, assuming the null hypothesis is true. This probability is called the p-value. If the p-value is less than a chosen significance level (α), the null hypothesis is rejected; otherwise, it is not rejected. The null hypothesis can be simple, specifying the population distribution completely, or composite, if it does not. The test is subject to two types of error: a Type I error occurs when the null hypothesis is rejected despite being true, with probability equal to α; a Type II error occurs when it is accepted despite the alternative being true. The concept originated from an earlier notion of statistical significance, where a sample was considered sufficiently inconsistent with the null hypothesis. The modern statistical hypothesis test added formal rigor by making the alternative hypothesis explicit. The acceptance region is the set of test statistic values for which the null hypothesis is not rejected, while the critical region is the set for which it is rejected. A conservative test ensures the true probability of a Type I error never exceeds the nominal significance level.
Reader's Guide
The null hypothesis is central to statistical hypothesis testing, providing a default position that the data must contradict to establish an effect. It is typically a statement of 'no effect' or 'no difference,' and the test is designed to assess the strength of evidence against it. The null hypothesis and alternative hypothesis are mutually exclusive. The test does not conclude that the null hypothesis is false, only that the data are inconsistent with it at a given significance level. Errors can occur: a Type I error rejects a true null hypothesis, with probability α, while a Type II error fails to reject a false null hypothesis, with probability β. The concept is used across many fields, including medical trials, where a null hypothesis might state that a drug has no effect. The null hypothesis can be simple (specifying the population distribution completely) or composite (not specifying it completely). Its use adds mathematical rigor and philosophical consistency to statistical inference.
Did You Know?
- The null hypothesis is typically a statement of 'no effect' or 'no difference'.
- If the p-value is less than the significance level α, the null hypothesis is rejected.
- A Type I error occurs when the null hypothesis is rejected despite being true.
- The null hypothesis and alternative hypothesis are mutually exclusive statements.
More in Probability & Statistics 1-24
Spotted an error? Know more?
This is a living reference — every entry is fact-audited, and reader corrections feed straight into our audit queue. Suggest an edit · See this site's audit record