Null Hypothesis (H₀)

The "nothing is going on" assumption that a hypothesis test tries to refute — e.g. "the means are equal".

Also known as: H₀, null hypothesis

The null hypothesis H₀ states the status quo: no effect, no deviation, nothing new. The test tries to refute it using the data — not to confirm it.

  • H₀: μ = μ₀ — the mean matches a target
  • H₀: μ₁ = μ₂ — two groups share the same mean
  • H₀: σ₁² = σ₂² — two groups have the same variance
  • H₀: data are normally distributed (normality tests)

See also

Used in

In the Algorithm Lab

Sources

  • Neyman, J. & Pearson, E. S. — On the Problem of the Most Efficient Tests, 1933
  • Wasserstein, R. L. & Lazar, N. A. — The ASA Statement on p-Values, 2016