Hypothesis test for the mean of a Normal distribution
Studying this course? Keep your progress in one place.
Key facts
- states the claim with an equals sign, about the population mean , never .
- If , then ; the standard error is .
- Reject if the -value is below the significance level, or the sample mean is in the critical region.
- For a two-tailed test, split the significance level between the tails.
- Conclude in two parts: the decision, and what it means in context. Never say is proved.
Hypotheses and significance
Whether sample data give enough evidence against a claim about a population mean.
A hypothesis test decides whether sample data give enough evidence against a claim. The null hypothesis states the claim, such as . The alternative hypothesis says how the parameter differs: or (one-tailed) or (two-tailed). Hypotheses are about , never the sample mean.
The significance level is the probability, assuming , below which we reject (commonly or ). The critical region is the set of results that lead to rejecting , and the p-value is the probability, assuming , of a result at least as extreme as observed.
A manufacturer claims bottles hold ml on average. A test checks whether they hold less. Which hypotheses are correct?
The sample mean
: the sample mean varies less than a single observation.
If and a random sample of size is taken, the sample mean has the distribution .
Its standard deviation, , is the standard error. It falls as rises. The test needs to be known, given or assumed.
Worked example
Under , and . State the distribution of .
- 1
The mean of is .
- 2
The standard error is .
- 3
So the variance is .
and . What is the standard error?
One-tailed test
State the hypotheses, write the distribution of under , find the -value, compare, and conclude in context.
- State , and the significance level.
- Write the distribution of assuming .
- Find the -value, or the critical value.
- Reject if the -value is below the significance level, or lies in the critical region.
- Write the conclusion in context.
A -value works too: for a one-tailed test, reject if (upper) or (lower).
- 1
Hypotheses
state H₀, H₁ and the level
- 2
Distribution
of X̄ assuming H₀
- 3
p-value
or the critical value
- 4
Compare
reject H₀ or not
- 5
Conclude
in context
Worked example
, at the level, with , and . Carry out the test.
- 1
Standard error , so under , .
- 2
.
- 3
, so reject .
- 4
Check: , and .
In a upper-tail test you find . What is the decision?
Two-tailed tests
The critical region is split between both tails: in each at the level, so .
If , the critical region is in both tails. At the level put in each tail, so the critical values of are . Alternatively double the one-tail probability to get the -value and compare it with . This is a two-tailed test.
Worked example
, , , , . Test at the level.
- 1
Standard error .
- 2
Critical region: or .
- 3
is not in the critical region.
- 4
Equivalently and .
- 5
At the level, , so would be rejected.
For a two-tailed test at the level, how much probability goes in each tail?
Conclusions and assumptions
Give the decision about and what it means in context; a larger sample gives stronger evidence.
Always write a two-part conclusion. If the -value is less than the significance level, reject : 'there is sufficient evidence at the level that the mean journey time is longer than minutes'. Otherwise: 'there is insufficient evidence'. Never say is proved true or false.
The same difference in sample mean gives stronger evidence with a larger : a mean of gives with but with . The test assumes the sample is random, the population is Normal, and is known, given or assumed.
at the level. Which conclusion is correct?
Try an exam question
The mass of a box of cereal is Normal with standard deviation g. A claim says the mean is g. A random sample of boxes has mean g. Test, at the significance level, whether the mean mass is less than g.
[4 marks]
- [1],
- [1]Under , , since
- [1], so do not reject
- [1]There is insufficient evidence at the level that the mean mass is less than g
That's the notes covered.
Carry on to the next subtopic.
Try a question
Assuming the mean is g, the distribution of the sample mean of bags is
1 mark, marked against the mark scheme
Assuming the mean is g, the distribution of the sample mean of bags is
1 mark, marked against the mark scheme