Hypothesis testing concepts

Hypothesis testing concepts

Spec 4.2–5 · Edexcel International A Level Further Maths: Revision notes

Studying this course? Keep your progress in one place.

Key facts

  • A hypothesis test assumes H0\mathrm{H}_0 and rejects it only if the result is very unlikely.
  • Hypotheses are about a parameter (pp or λ\lambda), and H0\mathrm{H}_0 gives a single value.
  • The critical region is where H0\mathrm{H}_0 is rejected; its probability is at most the significance level.
  • Two-tailed tests split the significance level between the tails.
  • Conclude in context: "sufficient" or "insufficient" evidence, never "proved".

What a test does

It decides whether sample data make a claim about a parameter unlikely enough to reject.

A hypothesis test uses sample data to decide whether a claim about a population parameter is believable. Assume the claim is true, work out how likely the observed result would be, and reject the claim only if the result would be very unlikely.

A test can show evidence for or against a claim, but it never proves it. The conclusion is worded as sufficient or insufficient evidence.

  1. 1

    State hypotheses

    H0\mathrm{H}_0 and H1\mathrm{H}_1

  2. 2

    Assume H₀ is true

    Find the distribution

  3. 3

    Find the critical region

    At the significance level

  4. 4

    Decide

    Is the result inside?

  5. 5

    Conclude

    In context

A hypothesis test

A test result is not in the critical region. Which conclusion is correct?

Null and alternative hypotheses

Both are statements about a parameter; H0\mathrm{H}_0 gives a single value.

The null hypothesis, H0\mathrm{H}_0, is the claim assumed to be true, and gives a single value of the parameter, for example H0:p=0.5\mathrm{H}_0:p=0.5. The alternative hypothesis, H1\mathrm{H}_1, says how the parameter differs: p>0.5p>0.5, p<0.5p<0.5 or p≠0.5p\ne0.5.

Define the parameter in context, such as "let pp be the probability that the coin lands heads".

00.10.20.30.40.50.60.70.80.91H₀: p = 0.5

Null hypothesis H0\mathrm{H}_0

  • The claim assumed true
  • A single value, e.g. p=0.5p=0.5

Alternative H1\mathrm{H}_1

  • How the parameter differs
  • p>0.5p>0.5, p<0.5p<0.5 or p≠0.5p\ne0.5

Which is a correct pair of hypotheses?

Critical region

The values of the test statistic for which H0\mathrm{H}_0 is rejected.

The test statistic is calculated from the sample, such as the number of heads XX; under H0\mathrm{H}_0 its distribution is known, such as X∼B(20,0.5)X\sim\mathrm{B}(20,0.5). The critical region is the set of values for which H0\mathrm{H}_0 is rejected, with edge the critical value. The significance level is the probability of falling in it when H0\mathrm{H}_0 is true.

Because the statistic is discrete, use the largest region with probability at most the level. The actual level is usually a little smaller.

00.050.10.150.20.250.0010.006012345678kP(X ≤ k)
P(X≤k)P(X\le k) for X∼B(20,0.5)X\sim\mathrm{B}(20,0.5): at 5%5\% the region is X≤5X\le5, since P(X≤6)=0.058P(X\le6)=0.058.

X∼B(20,0.5)X\sim\mathrm{B}(20,0.5). What is the lower-tail critical region at the 1%1\% level?

One-tailed and two-tailed

A one-tailed test has all the level in one tail; a two-tailed test splits it.

In a one-tailed test the alternative has a single direction, H1:p>p0\mathrm{H}_1:p>p_0 or p<p0p<p_0, used when the claim suggests a direction such as "the rate has fallen". In a two-tailed test the alternative is H1:p≠p0\mathrm{H}_1:p\ne p_0, and a 5%5\% test has 2.5%2.5\% in each tail.

For X∼B(20,0.5)X\sim\mathrm{B}(20,0.5) two-tailed at 5%5\%: P(X≤5)=0.0207<0.025P(X\le5)=0.0207<0.025 but P(X≤6)=0.0577>0.025P(X\le6)=0.0577>0.025, so the region is X≤5X\le5 or X≥15X\ge15, and the actual level is 2×0.0207=0.04142\times0.0207=0.0414.

00.050.10.150.00020.00110.00460.00460.0011234567891011121314151617XP(X = x)
Probabilities for X ~ B(20, 0.5) (X = 2 to 17 shown). The two-tailed 5% critical region is X ≤ 5 or X ≥ 15 (each tail below 0.025).

One-tailed

  • H1:p>p0\mathrm{H}_1:p>p_0 or p<p0p<p_0
  • All the level in one tail
  • "fallen", "biased towards"

Two-tailed

  • H1:p≠p0\mathrm{H}_1:p\ne p_0
  • Level split: 2.5% each tail
  • "has changed"

A die is tested to see whether it gives fewer sixes than expected, with H1:p<16\mathrm{H}_1:p<\frac16. This is a:

Concluding in context

Reject H0\mathrm{H}_0 if the result is in the critical region, then say what it means in context.

If the observed value is in the critical region, reject H0\mathrm{H}_0; otherwise do not reject it. Then write a conclusion in context that is not over-confident.

For example: "There is sufficient evidence, at the 5%5\% level, that the coin is biased towards heads." or "There is insufficient evidence that the pupil is not just guessing." A larger significance level makes rejection easier.

  1. 1

    Compare

    is the observed value in the critical region?

  2. 2

    Decide

    in the region: reject H₀; otherwise do not reject H₀

  3. 3

    Conclude in context

    say what it means, without being over-confident

Making a conclusion.

In the critical region

  • Reject H0\mathrm{H}_0
  • Sufficient evidence for mathrmH1mathrm{H}_1

Not in the critical region

  • Do not reject H0\mathrm{H}_0
  • Insufficient evidence for mathrmH1mathrm{H}_1

A result lies in the critical region at the 5%5\% level. What is the correct conclusion?

Refining models

A model is a claim about a parameter, so new data can test it, and it is refined if rejected.

A model such as "late arrivals are Poisson with mean 44" is a claim about a parameter, so it can be tested with new data. If the result is in the critical region, the model no longer fits and should be refined, for example by changing the mean.

If H0\mathrm{H}_0 is not rejected, there is no evidence against the model and it may be kept, although it has not been proved correct.

  1. 1

    Propose a model

    A claim about a parameter

  2. 2

    Collect data

    New observations

  3. 3

    Test H₀

    At a significance level

  4. 4

    Refine if rejected

    Change the parameter or model

Then back to step 1

Refining a model

The test data do not reject H0\mathrm{H}_0 for a Poisson model with mean 44. What can you say about the model?

Try an exam question

A coin is thought to be biased towards heads. It is tossed 2020 times and lands heads 1515 times. Test, at the 5%5\% significance level, whether the coin is biased towards heads. Use P(X≤5)=0.0207P(X\le5)=0.0207 and P(X≤6)=0.0577P(X\le6)=0.0577 for X∼B(20,0.5)X\sim\mathrm{B}(20,0.5).

[5 marks]

That's the notes covered.

Carry on to the next subtopic.

Try a question

Which pair of hypotheses should be used to test the suspicion?

1 mark, marked against the mark scheme

Revision notes