Choose your reading route
eduKateSG · Why Mathematics? · Confidence Intervals: Coverage, Margin of Error and Honest Uncertainty
Start with the question, audit the worked example, or jump to the assumptions and practice. Every chapter stays open for uninterrupted reading.
Understand the question
Define the quantity and follow the mechanism.Route 2 · Chapters 8–14
Reproduce the example
Audit the numbers one step at a time.Route 3 · Chapters 15–21
Check assumptions
See where design and calibration matter.Route 4 · Chapters 22–28
Explore applications
Connect the method to a concrete investigation.Route 5 · Chapters 29–34
Study and verify
Practise, inspect failure cases and read further.
Full chapter index · Practice and checking · Mathematics Learning Hub
Why is mathematics important when a single estimate needs an honest range of uncertainty? A confidence interval combines an estimate, its standard error and a critical value. Its confidence level describes the long-run coverage of the method under stated assumptions, not a probability attached to a fixed population value after the interval is observed.
Did You Know? A 95% interval procedure can miss the true value in about 1 out of 20 repetitions even when every assumption is satisfied.
Full chapter index
Chapters 1–7 · Understand the question
Chapters 8–14 · Reproduce the example
Chapters 15–21 · Check assumptions
Chapters 22–28 · Explore applications
Chapters 29–34 · Study and verify
Point estimates hide sampling variation. An interval reports values compatible with the estimate under a defined repeated-sampling procedure.
A symmetric normal-approximation interval is estimate ± z×SE. For 95% confidence, z is about 1.96.
ONS explains that across repeated samples, about 19 of 20 analogous 95% intervals would contain the true population value. Read the official uncertainty guide.
Suppose an estimated mean is 70 marks with SE=2 marks.
| Estimate | SE | Critical value |
|---|---|---|
| 70 | 2 | 1.96 |
Margin=1.96×2=3.92 marks.
The 95% interval is 70±3.92, or 66.08 to 73.92 marks.
| Margin | Lower | Upper |
|---|---|---|
| 3.92 | 66.08 | 73.92 |
Before sampling, the random interval has 95% long-run coverage under the model or design. After calculation, the endpoints are fixed; the procedure either covered the fixed target or did not.
Ninety-five percent is a performance rate, not a guarantee for this one interval.
Small effective samples may use a t critical value with design degrees of freedom. Replacing it automatically with 1.96 can make intervals too narrow.
Proportions near zero, ratios, odds and skewed estimators may need transformed, score, likelihood or bootstrap intervals.
A naive normal interval for a proportion can extend below zero or above one. Appropriate methods respect the parameter space.
Use design-aware SEs and degrees of freedom. A correct point weight with a naive SE still gives a misleading interval.
One 95% interval has its stated pointwise coverage. Looking at many intervals increases the chance at least one misses; multiplicity methods address a family of claims.
A confidence interval for a mean response differs from a prediction interval for a new individual, which includes individual variation.
A credible interval has a posterior-probability interpretation under a prior and likelihood. It should not be relabelled as a frequentist confidence interval.
A narrow interval around a tiny effect can be statistically precise but practically unimportant. A wide interval can include both beneficial and harmful values.
Larger n usually lowers SE and narrows the interval, with diminishing returns under the square-root law.
Coverage formulas usually exclude frame bias, nonresponse bias, measurement error and changing definitions unless explicitly modelled.
Check target, estimator, SE method, distributional approximation, degrees of freedom, bounds, multiplicity and practical thresholds.
Reproduce 66.08–73.92, simulate repeated 95% intervals, count coverage, test small n and compare a bounded proportion method.
Nominal 95% coverage can fail when assumptions, models, design variables or variance estimators are wrong.
Store estimate, SE, confidence level, critical value, margin, lower and upper bounds, method and degrees of freedom.
Simulate coverage, unit-test interval arithmetic, compare normal, t and bounded-proportion methods, and flag impossible bounds.
Scientists, engineers, clinicians and education analysts use intervals to communicate uncertainty.
Generate 1,000 samples from a known population, construct intervals and record the coverage rate and average width.
A 95% interval is not 95% of observations, it is not a guarantee, and overlapping intervals do not directly test a difference.
CHAPTER 27 OF 34 · Explore applications
27. Connection to recent mathematics articles
The standard-error guide supplies the uncertainty scale. The sample-size guide links width to planning.
Students should state the method’s repeated-sampling meaning. Parents can ask whether a school result’s interval is for the average or for an individual pupil.
- Calculate 3.92.
- Calculate both endpoints.
- Explain coverage.
- Choose t for small samples.
- Repair an impossible proportion interval.
### Is there a 95% frequentist probability the fixed target is inside? That is not the standard interpretation.
### Can intervals be asymmetric? Yes.
### Do wider intervals mean less precision? Yes.
### Do intervals include all bias? No.
Report target, estimate, units, confidence level, interval method, SE and design, critical value, degrees of freedom, bounds, multiplicity and practical thresholds.
Repeat the entire sampling and interval procedure many times; the confidence level is the fraction of intervals expected to cover the fixed target.
Confirm coverage language, design-aware SE, suitable critical value, parameter bounds, multiplicity, practical thresholds and explicit exclusions for nonsampling error.
Continue through the Mathematics Learning Hub. Recalculate one uncertainty measure and explain which decision it can and cannot support.
Extended practice laboratory
Practice case 1: Define target
Setup: Start from the article’s confidence-interval mechanism for case 1. Task: Change the named quantity, recompute the dependent result and compare with the original.
Check: Preserve the target population and sampling design, verify the uncertainty identity, state one diagnostic, record units and compare one defensible alternative with clear reasoning. Show the arithmetic, name the response and predictor units, and separate the numerical result from the claim it supports. Then vary one defensible assumption, calculate again and explain whether the conclusion changed because of the data, the fitted model or the validation purpose. Finish with one compact table or graph that could reveal a failure hidden by the summary.
Practice case 2: Set estimate
Setup: Start from the article’s confidence-interval mechanism for case 2. Task: Change the named quantity, recompute the dependent result and compare with the original.
Check: Preserve the target population and sampling design, verify the uncertainty identity, state one diagnostic, record units and compare one defensible alternative with clear reasoning. Show the arithmetic, name the response and predictor units, and separate the numerical result from the claim it supports. Then vary one defensible assumption, calculate again and explain whether the conclusion changed because of the data, the fitted model or the validation purpose. Finish with one compact table or graph that could reveal a failure hidden by the summary.
Practice case 3: Set SE
Setup: Start from the article’s confidence-interval mechanism for case 3. Task: Change the named quantity, recompute the dependent result and compare with the original.
Check: Preserve the target population and sampling design, verify the uncertainty identity, state one diagnostic, record units and compare one defensible alternative with clear reasoning. Show the arithmetic, name the response and predictor units, and separate the numerical result from the claim it supports. Then vary one defensible assumption, calculate again and explain whether the conclusion changed because of the data, the fitted model or the validation purpose. Finish with one compact table or graph that could reveal a failure hidden by the summary.
Practice case 4: Choose confidence
Setup: Start from the article’s confidence-interval mechanism for case 4. Task: Change the named quantity, recompute the dependent result and compare with the original.
Check: Preserve the target population and sampling design, verify the uncertainty identity, state one diagnostic, record units and compare one defensible alternative with clear reasoning. Show the arithmetic, name the response and predictor units, and separate the numerical result from the claim it supports. Then vary one defensible assumption, calculate again and explain whether the conclusion changed because of the data, the fitted model or the validation purpose. Finish with one compact table or graph that could reveal a failure hidden by the summary.
Practice case 5: Choose critical value
Setup: Start from the article’s confidence-interval mechanism for case 5. Task: Change the named quantity, recompute the dependent result and compare with the original.
Check: Preserve the target population and sampling design, verify the uncertainty identity, state one diagnostic, record units and compare one defensible alternative with clear reasoning. Show the arithmetic, name the response and predictor units, and separate the numerical result from the claim it supports. Then vary one defensible assumption, calculate again and explain whether the conclusion changed because of the data, the fitted model or the validation purpose. Finish with one compact table or graph that could reveal a failure hidden by the summary.
Practice case 6: Calculate margin
Setup: Start from the article’s confidence-interval mechanism for case 6. Task: Change the named quantity, recompute the dependent result and compare with the original.
Check: Preserve the target population and sampling design, verify the uncertainty identity, state one diagnostic, record units and compare one defensible alternative with clear reasoning. Show the arithmetic, name the response and predictor units, and separate the numerical result from the claim it supports. Then vary one defensible assumption, calculate again and explain whether the conclusion changed because of the data, the fitted model or the validation purpose. Finish with one compact table or graph that could reveal a failure hidden by the summary.
Practice case 7: Calculate lower
Setup: Start from the article’s confidence-interval mechanism for case 7. Task: Change the named quantity, recompute the dependent result and compare with the original.
Check: Preserve the target population and sampling design, verify the uncertainty identity, state one diagnostic, record units and compare one defensible alternative with clear reasoning. Show the arithmetic, name the response and predictor units, and separate the numerical result from the claim it supports. Then vary one defensible assumption, calculate again and explain whether the conclusion changed because of the data, the fitted model or the validation purpose. Finish with one compact table or graph that could reveal a failure hidden by the summary.
Practice case 8: Calculate upper
Setup: Start from the article’s confidence-interval mechanism for case 8. Task: Change the named quantity, recompute the dependent result and compare with the original.
Check: Preserve the target population and sampling design, verify the uncertainty identity, state one diagnostic, record units and compare one defensible alternative with clear reasoning. Show the arithmetic, name the response and predictor units, and separate the numerical result from the claim it supports. Then vary one defensible assumption, calculate again and explain whether the conclusion changed because of the data, the fitted model or the validation purpose. Finish with one compact table or graph that could reveal a failure hidden by the summary.
Practice case 9: Explain coverage
Setup: Start from the article’s confidence-interval mechanism for case 9. Task: Change the named quantity, recompute the dependent result and compare with the original.
Check: Preserve the target population and sampling design, verify the uncertainty identity, state one diagnostic, record units and compare one defensible alternative with clear reasoning. Show the arithmetic, name the response and predictor units, and separate the numerical result from the claim it supports. Then vary one defensible assumption, calculate again and explain whether the conclusion changed because of the data, the fitted model or the validation purpose. Finish with one compact table or graph that could reveal a failure hidden by the summary.
Practice case 10: Repeat sampling
Setup: Start from the article’s confidence-interval mechanism for case 10. Task: Change the named quantity, recompute the dependent result and compare with the original.
Check: Preserve the target population and sampling design, verify the uncertainty identity, state one diagnostic, record units and compare one defensible alternative with clear reasoning. Show the arithmetic, name the response and predictor units, and separate the numerical result from the claim it supports. Then vary one defensible assumption, calculate again and explain whether the conclusion changed because of the data, the fitted model or the validation purpose. Finish with one compact table or graph that could reveal a failure hidden by the summary.
Practice case 11: Use t value
Setup: Start from the article’s confidence-interval mechanism for case 11. Task: Change the named quantity, recompute the dependent result and compare with the original.
Check: Preserve the target population and sampling design, verify the uncertainty identity, state one diagnostic, record units and compare one defensible alternative with clear reasoning. Show the arithmetic, name the response and predictor units, and separate the numerical result from the claim it supports. Then vary one defensible assumption, calculate again and explain whether the conclusion changed because of the data, the fitted model or the validation purpose. Finish with one compact table or graph that could reveal a failure hidden by the summary.
Practice case 12: Add degrees freedom
Setup: Start from the article’s confidence-interval mechanism for case 12. Task: Change the named quantity, recompute the dependent result and compare with the original.
Check: Preserve the target population and sampling design, verify the uncertainty identity, state one diagnostic, record units and compare one defensible alternative with clear reasoning. Show the arithmetic, name the response and predictor units, and separate the numerical result from the claim it supports. Then vary one defensible assumption, calculate again and explain whether the conclusion changed because of the data, the fitted model or the validation purpose. Finish with one compact table or graph that could reveal a failure hidden by the summary.
Practice case 13: Use bounded proportion
Setup: Start from the article’s confidence-interval mechanism for case 13. Task: Change the named quantity, recompute the dependent result and compare with the original.
Check: Preserve the target population and sampling design, verify the uncertainty identity, state one diagnostic, record units and compare one defensible alternative with clear reasoning. Show the arithmetic, name the response and predictor units, and separate the numerical result from the claim it supports. Then vary one defensible assumption, calculate again and explain whether the conclusion changed because of the data, the fitted model or the validation purpose. Finish with one compact table or graph that could reveal a failure hidden by the summary.
Practice case 14: Use transformation
Setup: Start from the article’s confidence-interval mechanism for case 14. Task: Change the named quantity, recompute the dependent result and compare with the original.
Check: Preserve the target population and sampling design, verify the uncertainty identity, state one diagnostic, record units and compare one defensible alternative with clear reasoning. Show the arithmetic, name the response and predictor units, and separate the numerical result from the claim it supports. Then vary one defensible assumption, calculate again and explain whether the conclusion changed because of the data, the fitted model or the validation purpose. Finish with one compact table or graph that could reveal a failure hidden by the summary.
Practice case 15: Compare credible interval
Setup: Start from the article’s confidence-interval mechanism for case 15. Task: Change the named quantity, recompute the dependent result and compare with the original.
Check: Preserve the target population and sampling design, verify the uncertainty identity, state one diagnostic, record units and compare one defensible alternative with clear reasoning. Show the arithmetic, name the response and predictor units, and separate the numerical result from the claim it supports. Then vary one defensible assumption, calculate again and explain whether the conclusion changed because of the data, the fitted model or the validation purpose. Finish with one compact table or graph that could reveal a failure hidden by the summary.
Practice case 16: Add multiplicity
Setup: Start from the article’s confidence-interval mechanism for case 16. Task: Change the named quantity, recompute the dependent result and compare with the original.
Check: Preserve the target population and sampling design, verify the uncertainty identity, state one diagnostic, record units and compare one defensible alternative with clear reasoning. Show the arithmetic, name the response and predictor units, and separate the numerical result from the claim it supports. Then vary one defensible assumption, calculate again and explain whether the conclusion changed because of the data, the fitted model or the validation purpose. Finish with one compact table or graph that could reveal a failure hidden by the summary.
Practice case 17: Add practical threshold
Setup: Start from the article’s confidence-interval mechanism for case 17. Task: Change the named quantity, recompute the dependent result and compare with the original.
Check: Preserve the target population and sampling design, verify the uncertainty identity, state one diagnostic, record units and compare one defensible alternative with clear reasoning. Show the arithmetic, name the response and predictor units, and separate the numerical result from the claim it supports. Then vary one defensible assumption, calculate again and explain whether the conclusion changed because of the data, the fitted model or the validation purpose. Finish with one compact table or graph that could reveal a failure hidden by the summary.
Practice case 18: Change n
Setup: Start from the article’s confidence-interval mechanism for case 18. Task: Change the named quantity, recompute the dependent result and compare with the original.
Check: Preserve the target population and sampling design, verify the uncertainty identity, state one diagnostic, record units and compare one defensible alternative with clear reasoning. Show the arithmetic, name the response and predictor units, and separate the numerical result from the claim it supports. Then vary one defensible assumption, calculate again and explain whether the conclusion changed because of the data, the fitted model or the validation purpose. Finish with one compact table or graph that could reveal a failure hidden by the summary.
Practice case 19: Add design effect
Setup: Start from the article’s confidence-interval mechanism for case 19. Task: Change the named quantity, recompute the dependent result and compare with the original.
Check: Preserve the target population and sampling design, verify the uncertainty identity, state one diagnostic, record units and compare one defensible alternative with clear reasoning. Show the arithmetic, name the response and predictor units, and separate the numerical result from the claim it supports. Then vary one defensible assumption, calculate again and explain whether the conclusion changed because of the data, the fitted model or the validation purpose. Finish with one compact table or graph that could reveal a failure hidden by the summary.
Practice case 20: Add bias
Setup: Start from the article’s confidence-interval mechanism for case 20. Task: Change the named quantity, recompute the dependent result and compare with the original.
Check: Preserve the target population and sampling design, verify the uncertainty identity, state one diagnostic, record units and compare one defensible alternative with clear reasoning. Show the arithmetic, name the response and predictor units, and separate the numerical result from the claim it supports. Then vary one defensible assumption, calculate again and explain whether the conclusion changed because of the data, the fitted model or the validation purpose. Finish with one compact table or graph that could reveal a failure hidden by the summary.
Practice case 21: Audit method
Setup: Start from the article’s confidence-interval mechanism for case 21. Task: Change the named quantity, recompute the dependent result and compare with the original.
Check: Preserve the target population and sampling design, verify the uncertainty identity, state one diagnostic, record units and compare one defensible alternative with clear reasoning. Show the arithmetic, name the response and predictor units, and separate the numerical result from the claim it supports. Then vary one defensible assumption, calculate again and explain whether the conclusion changed because of the data, the fitted model or the validation purpose. Finish with one compact table or graph that could reveal a failure hidden by the summary.
Practice case 22: Audit language
Setup: Start from the article’s confidence-interval mechanism for case 22. Task: Change the named quantity, recompute the dependent result and compare with the original.
Check: Preserve the target population and sampling design, verify the uncertainty identity, state one diagnostic, record units and compare one defensible alternative with clear reasoning. Show the arithmetic, name the response and predictor units, and separate the numerical result from the claim it supports. Then vary one defensible assumption, calculate again and explain whether the conclusion changed because of the data, the fitted model or the validation purpose. Finish with one compact table or graph that could reveal a failure hidden by the summary.
Practice case 23: Audit bounds
Setup: Start from the article’s confidence-interval mechanism for case 23. Task: Change the named quantity, recompute the dependent result and compare with the original.
Check: Preserve the target population and sampling design, verify the uncertainty identity, state one diagnostic, record units and compare one defensible alternative with clear reasoning. Show the arithmetic, name the response and predictor units, and separate the numerical result from the claim it supports. Then vary one defensible assumption, calculate again and explain whether the conclusion changed because of the data, the fitted model or the validation purpose. Finish with one compact table or graph that could reveal a failure hidden by the summary.
Practice case 24: Name limits
Setup: Start from the article’s confidence-interval mechanism for case 24. Task: Change the named quantity, recompute the dependent result and compare with the original.
Check: Preserve the target population and sampling design, verify the uncertainty identity, state one diagnostic, record units and compare one defensible alternative with clear reasoning. Show the arithmetic, name the response and predictor units, and separate the numerical result from the claim it supports. Then vary one defensible assumption, calculate again and explain whether the conclusion changed because of the data, the fitted model or the validation purpose. Finish with one compact table or graph that could reveal a failure hidden by the summary.
Practice case 25: Write conclusion
Setup: Start from the article’s confidence-interval mechanism for case 25. Task: Change the named quantity, recompute the dependent result and compare with the original.
Check: Preserve the target population and sampling design, verify the uncertainty identity, state one diagnostic, record units and compare one defensible alternative with clear reasoning. Show the arithmetic, name the response and predictor units, and separate the numerical result from the claim it supports. Then vary one defensible assumption, calculate again and explain whether the conclusion changed because of the data, the fitted model or the validation purpose. Finish with one compact table or graph that could reveal a failure hidden by the summary.