Choose your reading route
eduKateSG · Why Mathematics? · Randomised Block Designs: Nuisance Variation and Within-Block Comparisons
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 29–42
Study and verify
Practise, inspect failure cases and read further.
Full chapter index · Practice and checking · Mathematics Learning Hub
Why is mathematics important when day, batch, classroom or participant differences add noise to an experiment? Randomised block designs compare treatments within homogeneous groups. Blocking removes known nuisance variation from the error term while randomisation protects each within-block comparison.
Did You Know? Pairing is the smallest complete block design: each participant or matched unit receives or represents both treatments. Precision improves when responses within a block are genuinely similar before treatment.
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 29–42 · Study and verify
- What students should transfer
- A spreadsheet workflow
- A coding workflow
- Applications and careers
- A student investigation
- Common misconceptions
- Connection to recent mathematics articles
- Guidance for students and parents
- Practice questions
- Frequently asked questions
- A reporting checklist
- Block what you know
- A final block audit
- Useful next reading
The treatment factor answers the research question. A block factor such as day, furnace run or baseline-matched pair affects response but is not the primary target.
Group units that are similar on the nuisance factor. Every treatment should appear within each complete block so treatment is compared under the same local conditions.
Randomly assign treatment labels separately inside each block. Blocking without randomisation can preserve systematic order bias.
NIST summarises the idea as blocking important controllable nuisance factors and randomising the rest. Read the NIST handbook.
Four blocks produce B-minus-A differences of 6, 2, 5 and 3 units. The treatment estimate is the mean of these within-block differences.
The differences sum to 16, so the paired estimate is 16/4=4 units.
The sample SD is sqrt(10/3)=1.8257 and the SE is 1.8257/sqrt(4)=0.9129.
With three degrees of freedom, t critical is about 3.182. The 95% interval is 4±3.182×0.9129, approximately 1.09 to 6.91.
| Block | B-A |
|---|---|
| 1 | 6 |
| 2 | 2 |
| 3 | 5 |
| 4 | 3 |
Between-block shifts cancel from each difference. If blocks explain substantial response variation, the residual error and treatment SE decrease.
An unhelpful block uses degrees of freedom but usually does not create major bias when the design and model are correct. Excessive tiny blocks can complicate execution.
A complete block model is response=overall mean+treatment effect+block effect+error. It assumes treatment differences do not systematically change across blocks.
| Mean | SD | SE | 95% t interval |
|---|---|---|---|
| 4 | 1.8257 | 0.9129 | 1.09 to 6.91 |
With one observation per treatment-block cell, interaction and error may not be separately estimable. Scientific judgement and replication matter.
When every block cannot contain every treatment, balanced incomplete block designs preserve planned connectivity. Analysis must use the actual incidence structure.
A failed observation breaks perfect balance. Use a model based on observed cells and report lost precision rather than inserting a convenient mean.
Both create local comparisons. In experiments, blocks describe groups of units or runs; stratification often describes assignment categories or analysis layers.
A continuous nuisance variable may be handled more efficiently in regression. Blocks are useful when local grouping is operationally natural.
A participant can act as a block when receiving several conditions. Period, carryover and within-person correlation then need extra modelling.
If all treatment A runs occur early and B late, treatment is confounded with time. Randomise treatment order within each batch.
Compare two teaching prompts within matched classes or days. Blocking on teacher or baseline attainment can reduce noise, but carryover and contamination must be managed.
Test four material dosages within each furnace run. Furnace variation is absorbed by block effects, leaving cleaner dosage contrasts.
Power depends on the SD of within-block differences, number of blocks, effect, alpha and missing-block risk. Pilot correlations can inform planning.
Blocks do not prevent observer expectations. Blind outcome assessment and standardise measurement when feasible.
Generate block effects, treatment effects and residual noise. Compare complete randomisation and blocking across strong and weak block correlations.
Plot block profiles, within-block differences, residuals and treatment order. Look for interaction, outliers and missing cells.
Reproduce mean 4, SD 1.8257, SE 0.9129 and interval 1.09–6.91. Add one missing observation and refit.
Blocking cannot adjust unmeasured post-treatment change, rescue confounded ordering or generalise beyond the blocks studied.
Report block definition, formation, within-block randomisation, treatment order, missing cells, paired estimate and block-aware interval.
Block a few strong nuisance factors that are known before assignment. Keep the design executable and randomise every remaining choice.
Identify variation you can control, compare like with like and preserve randomness. That reasoning applies far beyond formal experiments.
Use one row per unit with block, treatment, randomisation order, response, deviation and paired contrast.
Fit treatment and block effects, verify paired arithmetic and simulate missing cells and interaction.
Engineers, teachers, agricultural researchers and laboratory scientists use blocking to prevent nuisance variation from hiding real contrasts.
Test paper-airplane designs on several days, running every design once per day in random order. Compare blocked and unblocked SEs.
Blocking is not matching after outcomes, a block effect is not the treatment effect, and pairing helps only when pairs are meaningfully correlated.
The crossover guide covers participant blocks across periods. The factorial guide covers blocking in multi-factor designs.
Students should calculate every within-block difference before averaging. Parents can ask whether school comparisons account for teacher, class and day variation.
- Calculate four differences.
- Reproduce the t interval.
- Remove one cell.
- Add interaction.
- Compare with complete randomisation.
### What is a block? A homogeneous group used for local comparison.
### Must every treatment appear? In a complete block, yes.
### Is pairing blocking? Yes, with block size two.
### Can blocks be formed after outcomes? No.
Report treatment, experimental unit, nuisance factor, block construction, randomisation, balance, missing cells, model, block interaction, effect and uncertainty.
Known nuisance variation belongs in the design; unknown variation is handled through randomisation and replication.
Confirm each treatment was randomised within comparable blocks and the analysis uses the same block structure that generated the data.
Continue through the Mathematics Learning Hub. Recalculate the central example, identify the assumptions and explain which claim the result can and cannot support.
Extended practice laboratory
Practice case 1: Name treatment
Setup: Start from the article’s randomised block mechanism for case 1. Task: Change the named quantity, recompute the dependent result and compare with the original.
Check: Preserve the estimand and design, verify the statistical 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: Name nuisance factor
Setup: Start from the article’s randomised block mechanism for case 2. Task: Change the named quantity, recompute the dependent result and compare with the original.
Check: Preserve the estimand and design, verify the statistical 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: Create blocks
Setup: Start from the article’s randomised block mechanism for case 3. Task: Change the named quantity, recompute the dependent result and compare with the original.
Check: Preserve the estimand and design, verify the statistical 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: Randomise within block
Setup: Start from the article’s randomised block mechanism for case 4. Task: Change the named quantity, recompute the dependent result and compare with the original.
Check: Preserve the estimand and design, verify the statistical 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: Record outcomes
Setup: Start from the article’s randomised block mechanism for case 5. Task: Change the named quantity, recompute the dependent result and compare with the original.
Check: Preserve the estimand and design, verify the statistical 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 differences
Setup: Start from the article’s randomised block mechanism for case 6. Task: Change the named quantity, recompute the dependent result and compare with the original.
Check: Preserve the estimand and design, verify the statistical 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 mean difference
Setup: Start from the article’s randomised block mechanism for case 7. Task: Change the named quantity, recompute the dependent result and compare with the original.
Check: Preserve the estimand and design, verify the statistical 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 SD
Setup: Start from the article’s randomised block mechanism for case 8. Task: Change the named quantity, recompute the dependent result and compare with the original.
Check: Preserve the estimand and design, verify the statistical 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: Calculate SE
Setup: Start from the article’s randomised block mechanism for case 9. Task: Change the named quantity, recompute the dependent result and compare with the original.
Check: Preserve the estimand and design, verify the statistical 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: Calculate interval
Setup: Start from the article’s randomised block mechanism for case 10. Task: Change the named quantity, recompute the dependent result and compare with the original.
Check: Preserve the estimand and design, verify the statistical 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: Fit additive model
Setup: Start from the article’s randomised block mechanism for case 11. Task: Change the named quantity, recompute the dependent result and compare with the original.
Check: Preserve the estimand and design, verify the statistical 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: Test block effect
Setup: Start from the article’s randomised block mechanism for case 12. Task: Change the named quantity, recompute the dependent result and compare with the original.
Check: Preserve the estimand and design, verify the statistical 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: Add unequal block
Setup: Start from the article’s randomised block mechanism for case 13. Task: Change the named quantity, recompute the dependent result and compare with the original.
Check: Preserve the estimand and design, verify the statistical 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: Add missing cell
Setup: Start from the article’s randomised block mechanism for case 14. Task: Change the named quantity, recompute the dependent result and compare with the original.
Check: Preserve the estimand and design, verify the statistical 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: Add treatment-block interaction
Setup: Start from the article’s randomised block mechanism for case 15. Task: Change the named quantity, recompute the dependent result and compare with the original.
Check: Preserve the estimand and design, verify the statistical 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: Compare complete randomisation
Setup: Start from the article’s randomised block mechanism for case 16. Task: Change the named quantity, recompute the dependent result and compare with the original.
Check: Preserve the estimand and design, verify the statistical 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: Estimate variance reduction
Setup: Start from the article’s randomised block mechanism for case 17. Task: Change the named quantity, recompute the dependent result and compare with the original.
Check: Preserve the estimand and design, verify the statistical 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: Add repeated measures
Setup: Start from the article’s randomised block mechanism for case 18. Task: Change the named quantity, recompute the dependent result and compare with the original.
Check: Preserve the estimand and design, verify the statistical 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 batch drift
Setup: Start from the article’s randomised block mechanism for case 19. Task: Change the named quantity, recompute the dependent result and compare with the original.
Check: Preserve the estimand and design, verify the statistical 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: Check concealment
Setup: Start from the article’s randomised block mechanism for case 20. Task: Change the named quantity, recompute the dependent result and compare with the original.
Check: Preserve the estimand and design, verify the statistical 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: Simulate power
Setup: Start from the article’s randomised block mechanism for case 21. Task: Change the named quantity, recompute the dependent result and compare with the original.
Check: Preserve the estimand and design, verify the statistical 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: Inspect residuals
Setup: Start from the article’s randomised block mechanism for case 22. Task: Change the named quantity, recompute the dependent result and compare with the original.
Check: Preserve the estimand and design, verify the statistical 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 allocation
Setup: Start from the article’s randomised block mechanism for case 23. Task: Change the named quantity, recompute the dependent result and compare with the original.
Check: Preserve the estimand and design, verify the statistical 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 randomised block mechanism for case 24. Task: Change the named quantity, recompute the dependent result and compare with the original.
Check: Preserve the estimand and design, verify the statistical 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 randomised block mechanism for case 25. Task: Change the named quantity, recompute the dependent result and compare with the original.
Check: Preserve the estimand and design, verify the statistical 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.