Hypothesis test calculator

Test
Alternative hypothesis
Significance level

H₀: p = 0.4, H₁: p < 0.4

Reject H₀

p-value 0.0160 against 0.05.

Working

  1. H₀: p = 0.4. H₁: p < 0.4. One-tailed test at the 5% level.
  2. Under H₀, X ~ B(20, 0.4).
  3. P(X ≤ 3) = 0.0160.
  4. Critical region: X ≤ 3, with actual significance level 0.0160.
  5. 0.0160 ≤ 0.05: 3 is in the critical region, so reject H₀.
  6. There is sufficient evidence at the 5% level that the probability is less than 0.4.

Your result

Reject H₀. p-value 0.0160 against 0.05. Critical region: X ≤ 3.

Practise hypothesis tests

A-level Maths Pro

Get an A* or your money back

£129 once

Price rises to £199 on 1 January. Save £70 by joining before then.

At least 12 months' access

By paying you agree to our Terms and get access straight away, which ends your 14-day cancellation right.

No VAT is charged. Conditions

Testing H₀: p = 0.4 against H₁: p < 0.4 with 3 successes in 20 trials, a hypothesis test calculator finds P(X ≤ 3) = 0.0160, below 5%, so 3 is in the critical region X ≤ 3.

Questions people ask

How do you write the conclusion of a hypothesis test?

Say whether you reject H₀, then put it in context without certainty: there is (or is not) sufficient evidence at the 5% level that the probability of a defect has increased.

What is a critical region?

The set of results that would lead you to reject H₀. In a binomial test it is the values with a tail probability no bigger than the significance level.

What is the actual significance level?

The real probability of landing in the critical region when H₀ is true. For a binomial test it is usually below the nominal level, because X only takes whole numbers.

How does a two-tailed test differ?

Split the significance level between the two tails: at 10%, each tail gets 5%, and you compare your p-value with 5% on the side your result falls.

Keep going

How we work this out

What it does

Set up a binomial or normal-mean test, then get the p-value, the critical region and the conclusion, step by step.

One- and two-tailed tests at 1%, 2.5%, 5% and 10%: a binomial proportion, and the mean of a normal distribution with known standard deviation.

Method

  1. The p-value is the probability, under H₀, of a result at least as extreme as the one observed. A two-tailed test compares it with half the significance level on the side the result falls.
  2. The binomial critical region is the set of values whose cumulative tail probability is no more than the level in that tail, with the actual significance level shown.
  3. The normal-mean test uses z = (x̄ − μ) ÷ (σ ÷ √n).