Lesson 7 of 7 · 18 min

Using intervals to judge claims

Decide whether a claimed proportion is plausible from where it sits relative to a confidence interval, word the conclusion carefully, and model real problems with their assumptions.

What you will be able to do

  • Decide whether a claimed value of pp is supported by a confidence interval.
  • Judge claims such as "a majority" or "fewer than 1 in 4" with an interval.
  • Interpret an interval in context and state the assumptions behind it.
  • Solve unfamiliar problems with interval estimates, with and without technology.

Confidence intervals earn their keep when someone makes a claim. A council says 70% of residents support a new bikeway. A company says fewer than 5% of its products are faulty. A poll says one candidate is ahead. An interval from a random sample tells you which claims the data support.

Key idea

Inside or outside

  • Claimed value inside the interval: the claim is plausible; the data are consistent with it. That doesn't prove it: other values in the interval are also plausible.
  • Claimed value outside the interval: the data give evidence against the claim, at that confidence level. Say which way the truth seems to lie (higher or lower).

The council survey interval (0.628,0.692)(0.628, 0.692). A claim of 65% is plausible; the claim of 70% lies outside, above the interval.

Worked example

Does the survey support the council's claim?

  • Complex familiar
  • Technology active

A city council claims that 70% of residents support a new bikeway. A random sample of 850 residents finds that 561 support it.

(a) Calculate an approximate 95% confidence interval for the proportion of all residents who support the bikeway.

(b) Use your interval to evaluate the council's claim.

5 steps

Try it yourself first, then check the working one step at a time.

Method

  1. Calculate the interval.

  2. Inside: plausible. Outside: evidence against, in a stated direction.

  3. Conclude in context.

Watch out for

  • Saying "the claim is proven false": the interval gives evidence, at a stated confidence level, not proof.

A random sample gives a 95% confidence interval of (0.214,0.266)(0.214, 0.266) for the proportion of Australian adults who have donated blood. A blood service says one in four adults has donated. What do the data say about this claim?

Choose one answer

Key idea

Claims about a majority, or a threshold

For "more than half", compare the whole interval with 0.5:

  • entirely above 0.5: evidence of a majority;
  • containing 0.5: we can't conclude there is a majority (or that there isn't);
  • entirely below 0.5: evidence of a minority.

A poll with p^=0.53\hat{p} = 0.53 and n=1000n = 1000 gives the 95% interval (0.499,0.561)(0.499, 0.561). Even though 53% "is ahead", the interval includes 0.5, so the poll can't confirm a majority.

A 95% confidence interval for the proportion of voters in an electorate who support Candidate A is (0.512,0.568)(0.512, 0.568). Which conclusion is best supported?

Choose one answer

Common mistake

A large sample can't fix a biased one

The interval assumes a random sample. A voluntary online poll of 20 000 readers, or koalas brought into a wildlife hospital because they look sick, can give a narrow interval around the wrong value. The margin of error measures random sampling variation only, not bias.

Worked example

Modelling a koala disease survey

  • Complex unfamiliar
  • Technology active

Researchers test 180 randomly selected koalas in a South East Queensland region and find that 63 have chlamydia.

(a) Calculate an approximate 95% confidence interval for the proportion of koalas in the region with chlamydia.

(b) A news story claims that "at least half of the region's koalas are infected". Evaluate this claim.

(c) The researchers want a margin of error of at most 0.05 in next year's survey. Using this year's result, how many koalas should they test?

(d) A volunteer suggests saving money by testing koalas brought in to a wildlife hospital. Explain why this would be a problem.

5 steps

Try it yourself first, then check the working one step at a time.

Method

  1. Calculate and check the interval.

  2. Compare a threshold claim with the whole interval.

  3. Use a prior p^\hat{p} for the sample size, rounded up.

  4. Comment on randomness: the interval assumes a random sample.

Watch out for

  • Using the margin of error to "fix" a biased sample: it measures random variation only.

Exam tip

Writing an interpretation or evaluation

A full-marks interpretation names the confidence level, the population and the proportion in context, and uses the interval's numbers:

"We are 95% confident that between 62.8% and 69.2% of all residents of the city support the bikeway."

A full-marks evaluation then compares the claim with the interval and says what it means: "The council's 70% lies above the interval, so the survey gives evidence that support is lower than the council claims." Add a comment on assumptions (random sample, honest responses) if the question asks you to evaluate the reasonableness.

Summary

  • A claimed value inside the interval is plausible (consistent with the data), not proven.

  • A claimed value outside the interval is evidence against the claim; say whether the truth seems higher or lower.

  • For "a majority", the whole interval must lie above 0.5.

  • Interpret in context: confidence level, population, proportion, and the interval in the units of the question.

  • Intervals assume a random sample; a large biased sample still gives a misleading interval.

Finished the lesson?

Try the 2 quick checks above and the lesson completes itself, or mark it complete now.

Lesson 7 of 70% read

Save your progress with a free account