Chapter map

1. 1. Eligible comparison
2. 2. Concordance
3. 3. Discordance
4. 4. Ties
5. 5. Candidate one
6. 6. Difference
7. 7. Sum
8. 8. Gamma one
9. 9. Candidate two
10. 10. Gamma two
11. 11. Selection
12. 12. Perfect score
13. 13. Undefined
14. 14. Duplicates
15. 15. Scale
16. 16. Claim

Visit the How Mathematics Works hub for the wider series and its connections to education, reasoning, science, computing, design and careers.

---

1. Eligible comparison

The central mathematical idea is concrete. Pair one within distance with one between distance.

A concrete situation

Cluster ordering is tested directly.

What the mathematics does not license

Define sampling or enumeration. Keeping this boundary clear makes the result easier to audit and transfer.

Try it, check it, transfer it

Record rule. Then explain the result to someone who has not seen the formula. If the explanation needs a hidden assumption, add that assumption to the model record.

---

2. Concordance

Here the geometry becomes operational. Within distance is smaller.

A concrete situation

Add one to C.

What the mathematics does not license

Strict inequality excludes ties. A trustworthy workflow records the choice before seeing the final test result.

Try it, check it, transfer it

Count C. Then explain the result to someone who has not seen the formula. If the explanation needs a hidden assumption, add that assumption to the model record.

---

3. Discordance

This step deserves a careful pause. Within distance is larger.

A concrete situation

Add one to D.

What the mathematics does not license

Direction must not be reversed. Evidence becomes stronger when the limitation is tested instead of hidden.

Try it, check it, transfer it

Count D. Then explain the result to someone who has not seen the formula. If the explanation needs a hidden assumption, add that assumption to the model record.

---

4. Ties

The useful question is not only how to calculate, but what the calculation assumes. Distances are equal.

A concrete situation

Record separately under the usual gamma rule.

What the mathematics does not license

Ties do not enter C+D. Keeping this boundary clear makes the result easier to audit and transfer.

Try it, check it, transfer it

Count T. Then explain the result to someone who has not seen the formula. If the explanation needs a hidden assumption, add that assumption to the model record.

---

5. Candidate one

A small example makes the mechanism visible. Set C=40,D=10.

A concrete situation

Fifty non-tied comparisons enter.

What the mathematics does not license

Shared observations create dependence. A trustworthy workflow records the choice before seeing the final test result.

Try it, check it, transfer it

Record counts. Then explain the result to someone who has not seen the formula. If the explanation needs a hidden assumption, add that assumption to the model record.

---

6. Difference

This is where modelling choices begin to affect evidence. Compute 40-10.

A concrete situation

Numerator=30.

What the mathematics does not license

This is net ordering success. Evidence becomes stronger when the limitation is tested instead of hidden.

Try it, check it, transfer it

Show step. Then explain the result to someone who has not seen the formula. If the explanation needs a hidden assumption, add that assumption to the model record.

---

7. Sum

The central mathematical idea is concrete. Compute 40+10.

A concrete situation

Denominator=50.

What the mathematics does not license

It must be positive. Keeping this boundary clear makes the result easier to audit and transfer.

Try it, check it, transfer it

Show step. Then explain the result to someone who has not seen the formula. If the explanation needs a hidden assumption, add that assumption to the model record.

---

8. Gamma one

Here the geometry becomes operational. Compute 30/50.

A concrete situation

Gamma=.60.

What the mathematics does not license

It measures order, not margin. A trustworthy workflow records the choice before seeing the final test result.

Try it, check it, transfer it

Record value. Then explain the result to someone who has not seen the formula. If the explanation needs a hidden assumption, add that assumption to the model record.

---

9. Candidate two

This step deserves a careful pause. Set C=32,D=18.

A concrete situation

Net difference is 14.

What the mathematics does not license

Use identical rules. Evidence becomes stronger when the limitation is tested instead of hidden.

Try it, check it, transfer it

Record counts. Then explain the result to someone who has not seen the formula. If the explanation needs a hidden assumption, add that assumption to the model record.

---

10. Gamma two

The useful question is not only how to calculate, but what the calculation assumes. Compute 14/50.

A concrete situation

Gamma=.28.

What the mathematics does not license

Do not compare unequal conventions. Keeping this boundary clear makes the result easier to audit and transfer.

Try it, check it, transfer it

Record value. Then explain the result to someone who has not seen the formula. If the explanation needs a hidden assumption, add that assumption to the model record.

---

11. Selection

A small example makes the mechanism visible. Compare .60 with .28.

A concrete situation

Maximum rule prefers candidate one.

What the mathematics does not license

Search range affects the winner. A trustworthy workflow records the choice before seeing the final test result.

Try it, check it, transfer it

State result. Then explain the result to someone who has not seen the formula. If the explanation needs a hidden assumption, add that assumption to the model record.

---

12. Perfect score

This is where modelling choices begin to affect evidence. Set D=0 and C>0.

A concrete situation

Gamma=1.

What the mathematics does not license

This does not prove stability. Evidence becomes stronger when the limitation is tested instead of hidden.

Try it, check it, transfer it

Check resamples. Then explain the result to someone who has not seen the formula. If the explanation needs a hidden assumption, add that assumption to the model record.

---

13. Undefined

The central mathematical idea is concrete. Set C=D=0.

A concrete situation

Gamma is undefined.

What the mathematics does not license

All eligible comparisons may be tied. Keeping this boundary clear makes the result easier to audit and transfer.

Try it, check it, transfer it

Flag failure. Then explain the result to someone who has not seen the formula. If the explanation needs a hidden assumption, add that assumption to the model record.

---

14. Duplicates

Here the geometry becomes operational. Add repeated observations.

A concrete situation

Tie count can surge.

What the mathematics does not license

Report duplicates and tie policy. A trustworthy workflow records the choice before seeing the final test result.

Try it, check it, transfer it

Audit. Then explain the result to someone who has not seen the formula. If the explanation needs a hidden assumption, add that assumption to the model record.

---

15. Scale

This step deserves a careful pause. Rescale one variable.

A concrete situation

Distance ranks can reorder.

What the mathematics does not license

Preprocessing is part of the evidence. Evidence becomes stronger when the limitation is tested instead of hidden.

Try it, check it, transfer it

Recompute. Then explain the result to someone who has not seen the formula. If the explanation needs a hidden assumption, add that assumption to the model record.

---

16. Claim

The useful question is not only how to calculate, but what the calculation assumes. Name C,D,T,metric,scaling,range and maximum rule.

A concrete situation

Candidate one has stronger ordering.

What the mathematics does not license

Do not claim universal categories. Keeping this boundary clear makes the result easier to audit and transfer.

Try it, check it, transfer it

Bound conclusion. Then explain the result to someone who has not seen the formula. If the explanation needs a hidden assumption, add that assumption to the model record.

---

A seven-day practice lab

1. Day 1: define the decision, observation unit, X variables and Y or reference condition.
2. Day 2: reproduce the smallest worked example and check every dimension, unit and sign.
3. Day 3: build a transparent baseline and write what it assumes.
4. Day 4: fit the main method inside a training-only pipeline; record every parameter.
5. Day 5: challenge one assumption with a deliberately difficult dataset or operating regime.
6. Day 6: compare validation metrics with the practical workload created by errors.
7. Day 7: present one supported conclusion, one unresolved uncertainty and one safe next experiment.

Questions for a student, parent or mentor

• Which part of the model is a mathematical definition, and which part is a choice?
• What small calculation would catch a sign, scale or dimension error?
• What evidence comes from training, validation and the untouched test?
• What benign condition could look like a fault or prediction failure?
• How would the answer change for another response, asset, school project or time period?
• What should a human review before acting on the score?

A method-specific audit before deployment

The Baker–Hubert gamma index turns a clustering into ordered comparisons between distances. Take one pair of observations from the same cluster and one pair from different clusters. The comparison is concordant when the within-cluster distance is smaller, discordant when it is larger and tied when the distances are equal. If C and D count concordant and discordant comparisons, gamma=(C-D)/(C+D).

For a candidate partition with C=40 and D=10, gamma=(40-10)/(40+10)=30/50=0.60. A second candidate with C=32 and D=18 has gamma=(32-18)/50=0.28. Under the maximum-gamma rule, the first partition is preferred because its within-versus-between ordering is more consistently correct.

The score ranges from -1 to 1 when C+D is positive. A value of 1 means every eligible non-tied comparison places the within distance below the between distance. A value of 0 means concordance and discordance balance. Negative values mean reversals are more common. Gamma measures ordering, not the size of the distance gaps.

Ties are excluded from the usual denominator, which can make a small set of non-tied comparisons look decisive. Comparisons share observations and are not independent Bernoulli trials. Distance choice, scaling and duplicate points matter, and maximising gamma across many cluster counts introduces selection. Report C, D, tie count, eligible-comparison rule, metric and candidate range.

---

Frequently asked questions

What is Baker–Hubert gamma?

The difference between concordant and discordant within-versus-between distance comparisons divided by their sum.

Which direction is better?

Higher gamma is preferred under the maximum rule.

What happens to tied distances?

The usual gamma denominator excludes them, so their count should be reported separately.

Does gamma measure the size of separation?

No. It measures ordering; small and large correct margins each count once.

Can comparisons be treated as independent?

No. Many comparisons share observations and distances.

What must be reported?

C, D, tie count, eligible-comparison rule, metric, scaling, candidate range, algorithm and maximum rule.

---

Primary source and fact-check note

Frank B. Baker and Lawrence J. Hubert, “Measuring the Power of Hierarchical Cluster Analysis,” Journal of the American Statistical Association 70(349), 31–38 (1975), DOI 10.1080/01621459.1975.10480256. Definitions and core claims in this article were checked against this primary publication. Calculations shown here were independently recomputed. Examples are educational illustrations; they are not reported industrial performance results.

Useful next reading

• G-plus discordance index
• Frey distance-change ratio
• TraceW within-cluster error
• How AI finds patterns without labels
• Bukit Timah Mathematics learning routes

Keep the options open. Mathematics education supports clearer questions, transferable problem-solving skills and informed choices across science, engineering, computing, AI, statistics, design and many other careers. It does not guarantee intelligence, income, admission or employment; those outcomes depend on far more than one subject or model.

Extended laboratory: cluster-validity evidence and responsible use

Data audit

Freeze observations, variables, units, exclusions, missing-data rules, weights, standardisation, sample size and seeds.

Hierarchy audit

Record clustering method, linkage, distance, adjacent levels, candidate range and tie handling.

Distance audit

Define within and between pairs, centroids, squared errors, dissimilarities, means and comparison eligibility.

Sequence audit

Show parent and child quantities, raw scores, signed changes, counts, denominators, probabilities and candidate labels.

Convention audit

Name exponent, parentheses, sign, tie, normalisation, threshold, alpha, scan-order and equality conventions.

Uncertainty audit

Repeat samples, feature scales, distances, linkages, starts, candidate ranges and plausible implementations.

Meaning audit

Compare internal numerical evidence with stability and domain evidence without assuming natural categories.

Final claim

State exact split or distance-order evidence and stop before causal, semantic, career or guaranteed-performance claims.

Mechanism workshop: calculate, challenge and transfer

Workshop 1: Eligible comparison

Begin with this mechanism: Pair one within distance with one between distance. Recreate the concrete case: Cluster ordering is tested directly. Write eligible comparisons, C, D, tie count, numerator, denominator, gamma values and chosen candidate with identifiers and assumptions. Calculate the smallest non-trivial example by hand and reproduce it independently. Challenge this limitation: Define sampling or enumeration. Complete the transfer step: Record rule. Change exactly one observation, variable, scale, distance, cluster count, hierarchy level, W value, parent size, dimension, mean distance, concordance count, discordance count, tie policy, denominator, alpha or seed. Explain which mathematical object changes, which score moves and which interpretation must weaken. Keep compactness, separation, hierarchy, stability, uncertainty and practical usefulness on separate notebook lines. Explain once to a student and once to a parent. Add a counterexample involving a constant variable, duplicate, outlier, zero denominator, tiny denominator, tied distance, missing pair or ambiguous merge. State the extra evidence needed to distinguish explanations. Finish with one supported sentence, one unsupported sentence and one useful next measurement.

Workshop 2: Concordance

Begin with this mechanism: Within distance is smaller. Recreate the concrete case: Add one to C. Write eligible comparisons, C, D, tie count, numerator, denominator, gamma values and chosen candidate with identifiers and assumptions. Calculate the smallest non-trivial example by hand and reproduce it independently. Challenge this limitation: Strict inequality excludes ties. Complete the transfer step: Count C. Change exactly one observation, variable, scale, distance, cluster count, hierarchy level, W value, parent size, dimension, mean distance, concordance count, discordance count, tie policy, denominator, alpha or seed. Explain which mathematical object changes, which score moves and which interpretation must weaken. Keep compactness, separation, hierarchy, stability, uncertainty and practical usefulness on separate notebook lines. Explain once to a student and once to a parent. Add a counterexample involving a constant variable, duplicate, outlier, zero denominator, tiny denominator, tied distance, missing pair or ambiguous merge. State the extra evidence needed to distinguish explanations. Finish with one supported sentence, one unsupported sentence and one useful next measurement.

Workshop 3: Discordance

Begin with this mechanism: Within distance is larger. Recreate the concrete case: Add one to D. Write eligible comparisons, C, D, tie count, numerator, denominator, gamma values and chosen candidate with identifiers and assumptions. Calculate the smallest non-trivial example by hand and reproduce it independently. Challenge this limitation: Direction must not be reversed. Complete the transfer step: Count D. Change exactly one observation, variable, scale, distance, cluster count, hierarchy level, W value, parent size, dimension, mean distance, concordance count, discordance count, tie policy, denominator, alpha or seed. Explain which mathematical object changes, which score moves and which interpretation must weaken. Keep compactness, separation, hierarchy, stability, uncertainty and practical usefulness on separate notebook lines. Explain once to a student and once to a parent. Add a counterexample involving a constant variable, duplicate, outlier, zero denominator, tiny denominator, tied distance, missing pair or ambiguous merge. State the extra evidence needed to distinguish explanations. Finish with one supported sentence, one unsupported sentence and one useful next measurement.

Workshop 4: Ties

Begin with this mechanism: Distances are equal. Recreate the concrete case: Record separately under the usual gamma rule. Write eligible comparisons, C, D, tie count, numerator, denominator, gamma values and chosen candidate with identifiers and assumptions. Calculate the smallest non-trivial example by hand and reproduce it independently. Challenge this limitation: Ties do not enter C+D. Complete the transfer step: Count T. Change exactly one observation, variable, scale, distance, cluster count, hierarchy level, W value, parent size, dimension, mean distance, concordance count, discordance count, tie policy, denominator, alpha or seed. Explain which mathematical object changes, which score moves and which interpretation must weaken. Keep compactness, separation, hierarchy, stability, uncertainty and practical usefulness on separate notebook lines. Explain once to a student and once to a parent. Add a counterexample involving a constant variable, duplicate, outlier, zero denominator, tiny denominator, tied distance, missing pair or ambiguous merge. State the extra evidence needed to distinguish explanations. Finish with one supported sentence, one unsupported sentence and one useful next measurement.

Workshop 5: Candidate one

Begin with this mechanism: Set C=40,D=10. Recreate the concrete case: Fifty non-tied comparisons enter. Write eligible comparisons, C, D, tie count, numerator, denominator, gamma values and chosen candidate with identifiers and assumptions. Calculate the smallest non-trivial example by hand and reproduce it independently. Challenge this limitation: Shared observations create dependence. Complete the transfer step: Record counts. Change exactly one observation, variable, scale, distance, cluster count, hierarchy level, W value, parent size, dimension, mean distance, concordance count, discordance count, tie policy, denominator, alpha or seed. Explain which mathematical object changes, which score moves and which interpretation must weaken. Keep compactness, separation, hierarchy, stability, uncertainty and practical usefulness on separate notebook lines. Explain once to a student and once to a parent. Add a counterexample involving a constant variable, duplicate, outlier, zero denominator, tiny denominator, tied distance, missing pair or ambiguous merge. State the extra evidence needed to distinguish explanations. Finish with one supported sentence, one unsupported sentence and one useful next measurement.

Workshop 6: Difference

Begin with this mechanism: Compute 40-10. Recreate the concrete case: Numerator=30. Write eligible comparisons, C, D, tie count, numerator, denominator, gamma values and chosen candidate with identifiers and assumptions. Calculate the smallest non-trivial example by hand and reproduce it independently. Challenge this limitation: This is net ordering success. Complete the transfer step: Show step. Change exactly one observation, variable, scale, distance, cluster count, hierarchy level, W value, parent size, dimension, mean distance, concordance count, discordance count, tie policy, denominator, alpha or seed. Explain which mathematical object changes, which score moves and which interpretation must weaken. Keep compactness, separation, hierarchy, stability, uncertainty and practical usefulness on separate notebook lines. Explain once to a student and once to a parent. Add a counterexample involving a constant variable, duplicate, outlier, zero denominator, tiny denominator, tied distance, missing pair or ambiguous merge. State the extra evidence needed to distinguish explanations. Finish with one supported sentence, one unsupported sentence and one useful next measurement.

Workshop 7: Sum

Begin with this mechanism: Compute 40+10. Recreate the concrete case: Denominator=50. Write eligible comparisons, C, D, tie count, numerator, denominator, gamma values and chosen candidate with identifiers and assumptions. Calculate the smallest non-trivial example by hand and reproduce it independently. Challenge this limitation: It must be positive. Complete the transfer step: Show step. Change exactly one observation, variable, scale, distance, cluster count, hierarchy level, W value, parent size, dimension, mean distance, concordance count, discordance count, tie policy, denominator, alpha or seed. Explain which mathematical object changes, which score moves and which interpretation must weaken. Keep compactness, separation, hierarchy, stability, uncertainty and practical usefulness on separate notebook lines. Explain once to a student and once to a parent. Add a counterexample involving a constant variable, duplicate, outlier, zero denominator, tiny denominator, tied distance, missing pair or ambiguous merge. State the extra evidence needed to distinguish explanations. Finish with one supported sentence, one unsupported sentence and one useful next measurement.

Workshop 8: Gamma one

Begin with this mechanism: Compute 30/50. Recreate the concrete case: Gamma=.60. Write eligible comparisons, C, D, tie count, numerator, denominator, gamma values and chosen candidate with identifiers and assumptions. Calculate the smallest non-trivial example by hand and reproduce it independently. Challenge this limitation: It measures order, not margin. Complete the transfer step: Record value. Change exactly one observation, variable, scale, distance, cluster count, hierarchy level, W value, parent size, dimension, mean distance, concordance count, discordance count, tie policy, denominator, alpha or seed. Explain which mathematical object changes, which score moves and which interpretation must weaken. Keep compactness, separation, hierarchy, stability, uncertainty and practical usefulness on separate notebook lines. Explain once to a student and once to a parent. Add a counterexample involving a constant variable, duplicate, outlier, zero denominator, tiny denominator, tied distance, missing pair or ambiguous merge. State the extra evidence needed to distinguish explanations. Finish with one supported sentence, one unsupported sentence and one useful next measurement.

Workshop 9: Candidate two

Begin with this mechanism: Set C=32,D=18. Recreate the concrete case: Net difference is 14. Write eligible comparisons, C, D, tie count, numerator, denominator, gamma values and chosen candidate with identifiers and assumptions. Calculate the smallest non-trivial example by hand and reproduce it independently. Challenge this limitation: Use identical rules. Complete the transfer step: Record counts. Change exactly one observation, variable, scale, distance, cluster count, hierarchy level, W value, parent size, dimension, mean distance, concordance count, discordance count, tie policy, denominator, alpha or seed. Explain which mathematical object changes, which score moves and which interpretation must weaken. Keep compactness, separation, hierarchy, stability, uncertainty and practical usefulness on separate notebook lines. Explain once to a student and once to a parent. Add a counterexample involving a constant variable, duplicate, outlier, zero denominator, tiny denominator, tied distance, missing pair or ambiguous merge. State the extra evidence needed to distinguish explanations. Finish with one supported sentence, one unsupported sentence and one useful next measurement.

Workshop 10: Gamma two

Begin with this mechanism: Compute 14/50. Recreate the concrete case: Gamma=.28. Write eligible comparisons, C, D, tie count, numerator, denominator, gamma values and chosen candidate with identifiers and assumptions. Calculate the smallest non-trivial example by hand and reproduce it independently. Challenge this limitation: Do not compare unequal conventions. Complete the transfer step: Record value. Change exactly one observation, variable, scale, distance, cluster count, hierarchy level, W value, parent size, dimension, mean distance, concordance count, discordance count, tie policy, denominator, alpha or seed. Explain which mathematical object changes, which score moves and which interpretation must weaken. Keep compactness, separation, hierarchy, stability, uncertainty and practical usefulness on separate notebook lines. Explain once to a student and once to a parent. Add a counterexample involving a constant variable, duplicate, outlier, zero denominator, tiny denominator, tied distance, missing pair or ambiguous merge. State the extra evidence needed to distinguish explanations. Finish with one supported sentence, one unsupported sentence and one useful next measurement.

Workshop 11: Selection

Begin with this mechanism: Compare .60 with .28. Recreate the concrete case: Maximum rule prefers candidate one. Write eligible comparisons, C, D, tie count, numerator, denominator, gamma values and chosen candidate with identifiers and assumptions. Calculate the smallest non-trivial example by hand and reproduce it independently. Challenge this limitation: Search range affects the winner. Complete the transfer step: State result. Change exactly one observation, variable, scale, distance, cluster count, hierarchy level, W value, parent size, dimension, mean distance, concordance count, discordance count, tie policy, denominator, alpha or seed. Explain which mathematical object changes, which score moves and which interpretation must weaken. Keep compactness, separation, hierarchy, stability, uncertainty and practical usefulness on separate notebook lines. Explain once to a student and once to a parent. Add a counterexample involving a constant variable, duplicate, outlier, zero denominator, tiny denominator, tied distance, missing pair or ambiguous merge. State the extra evidence needed to distinguish explanations. Finish with one supported sentence, one unsupported sentence and one useful next measurement.

Workshop 12: Perfect score

Begin with this mechanism: Set D=0 and C>0. Recreate the concrete case: Gamma=1. Write eligible comparisons, C, D, tie count, numerator, denominator, gamma values and chosen candidate with identifiers and assumptions. Calculate the smallest non-trivial example by hand and reproduce it independently. Challenge this limitation: This does not prove stability. Complete the transfer step: Check resamples. Change exactly one observation, variable, scale, distance, cluster count, hierarchy level, W value, parent size, dimension, mean distance, concordance count, discordance count, tie policy, denominator, alpha or seed. Explain which mathematical object changes, which score moves and which interpretation must weaken. Keep compactness, separation, hierarchy, stability, uncertainty and practical usefulness on separate notebook lines. Explain once to a student and once to a parent. Add a counterexample involving a constant variable, duplicate, outlier, zero denominator, tiny denominator, tied distance, missing pair or ambiguous merge. State the extra evidence needed to distinguish explanations. Finish with one supported sentence, one unsupported sentence and one useful next measurement.

Workshop 13: Undefined

Begin with this mechanism: Set C=D=0. Recreate the concrete case: Gamma is undefined. Write eligible comparisons, C, D, tie count, numerator, denominator, gamma values and chosen candidate with identifiers and assumptions. Calculate the smallest non-trivial example by hand and reproduce it independently. Challenge this limitation: All eligible comparisons may be tied. Complete the transfer step: Flag failure. Change exactly one observation, variable, scale, distance, cluster count, hierarchy level, W value, parent size, dimension, mean distance, concordance count, discordance count, tie policy, denominator, alpha or seed. Explain which mathematical object changes, which score moves and which interpretation must weaken. Keep compactness, separation, hierarchy, stability, uncertainty and practical usefulness on separate notebook lines. Explain once to a student and once to a parent. Add a counterexample involving a constant variable, duplicate, outlier, zero denominator, tiny denominator, tied distance, missing pair or ambiguous merge. State the extra evidence needed to distinguish explanations. Finish with one supported sentence, one unsupported sentence and one useful next measurement.

Workshop 14: Duplicates

Begin with this mechanism: Add repeated observations. Recreate the concrete case: Tie count can surge. Write eligible comparisons, C, D, tie count, numerator, denominator, gamma values and chosen candidate with identifiers and assumptions. Calculate the smallest non-trivial example by hand and reproduce it independently. Challenge this limitation: Report duplicates and tie policy. Complete the transfer step: Audit. Change exactly one observation, variable, scale, distance, cluster count, hierarchy level, W value, parent size, dimension, mean distance, concordance count, discordance count, tie policy, denominator, alpha or seed. Explain which mathematical object changes, which score moves and which interpretation must weaken. Keep compactness, separation, hierarchy, stability, uncertainty and practical usefulness on separate notebook lines. Explain once to a student and once to a parent. Add a counterexample involving a constant variable, duplicate, outlier, zero denominator, tiny denominator, tied distance, missing pair or ambiguous merge. State the extra evidence needed to distinguish explanations. Finish with one supported sentence, one unsupported sentence and one useful next measurement.

Workshop 15: Scale

Begin with this mechanism: Rescale one variable. Recreate the concrete case: Distance ranks can reorder. Write eligible comparisons, C, D, tie count, numerator, denominator, gamma values and chosen candidate with identifiers and assumptions. Calculate the smallest non-trivial example by hand and reproduce it independently. Challenge this limitation: Preprocessing is part of the evidence. Complete the transfer step: Recompute. Change exactly one observation, variable, scale, distance, cluster count, hierarchy level, W value, parent size, dimension, mean distance, concordance count, discordance count, tie policy, denominator, alpha or seed. Explain which mathematical object changes, which score moves and which interpretation must weaken. Keep compactness, separation, hierarchy, stability, uncertainty and practical usefulness on separate notebook lines. Explain once to a student and once to a parent. Add a counterexample involving a constant variable, duplicate, outlier, zero denominator, tiny denominator, tied distance, missing pair or ambiguous merge. State the extra evidence needed to distinguish explanations. Finish with one supported sentence, one unsupported sentence and one useful next measurement.

Workshop 16: Claim

Begin with this mechanism: Name C,D,T,metric,scaling,range and maximum rule. Recreate the concrete case: Candidate one has stronger ordering. Write eligible comparisons, C, D, tie count, numerator, denominator, gamma values and chosen candidate with identifiers and assumptions. Calculate the smallest non-trivial example by hand and reproduce it independently. Challenge this limitation: Do not claim universal categories. Complete the transfer step: Bound conclusion. Change exactly one observation, variable, scale, distance, cluster count, hierarchy level, W value, parent size, dimension, mean distance, concordance count, discordance count, tie policy, denominator, alpha or seed. Explain which mathematical object changes, which score moves and which interpretation must weaken. Keep compactness, separation, hierarchy, stability, uncertainty and practical usefulness on separate notebook lines. Explain once to a student and once to a parent. Add a counterexample involving a constant variable, duplicate, outlier, zero denominator, tiny denominator, tied distance, missing pair or ambiguous merge. State the extra evidence needed to distinguish explanations. Finish with one supported sentence, one unsupported sentence and one useful next measurement.