Choose your reading route
eduKateSG · Why Mathematics? · Weighted Mean, Unequal Importance and Frequency Averages
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 4–7
Reproduce the example
Audit the numbers one step at a time.Route 3 · Chapters 8–13
Check assumptions
See where design and calibration matter.Route 4 · Chapters 14–18
Explore applications
Connect the method to a concrete investigation.Route 5 · Chapters 19–27
Study and verify
Practise, inspect failure cases and read further.
Full chapter index · Practice and checking · Mathematics Learning Hub
Why is mathematics important when observations do not contribute equally? A weighted mean multiplies each value by a declared weight, adds the products and divides by the total weight. The arithmetic is simple; the responsibility lies in explaining what the weights mean.
Weighted averages appear in school assessment, price indices, survey estimation, laboratory combinations and machine learning. They can represent frequency, exposure, reliability or policy choice, and those meanings are not interchangeable.
Did you know? Doubling every weight leaves the weighted mean unchanged. Only the relative pattern of non-zero weights matters for the centre.
Full chapter index
Chapters 1–3 · Understand the question
Chapters 4–7 · Reproduce the example
Chapters 8–13 · Check assumptions
Chapters 14–18 · Explore applications
Chapters 19–27 · Study and verify
An ordinary mean gives each recorded value one equal vote. That is appropriate when observations are equally represented and equally relevant to the target.
A weighted mean changes those contributions according to an explicit design.
NIST gives x-bar-w = sum(w_i x_i) divided by sum(w_i). Every value and weight must be paired correctly.
The denominator is the weight sum, not automatically the number of rows.
| Values | Weights | Products |
|---|---|---|
| 60,80,90 | 0.2,0.3,0.5 | 12,24,45 |
Suppose three components have scores 60,80 and90 with weights 0.2,0.3 and0.5. The weighted products are12,24 and45.
Their sum is81, and the weights sum to1, so the weighted mean is81.
| Product sum | Weight sum | Weighted mean |
|---|---|---|
| 81 | 1 | 81/1 = 81 |
Take the same relative weights as2,3 and5. Products are120,240 and450, summing810. Divide by weight sum10 to get81 again.
Normalising first is convenient but not required.
If value2 occurs three times and value5 occurs once, the weighted mean is (3×2+1×5)/(3+1)=11/4=2.75.
This equals the ordinary mean of the expanded list2,2,2,5.
In measurement synthesis, more precise estimates may receive greater weight, often related to inverse variance under a specified model.
That rule carries statistical assumptions; it is not the same as frequency weighting.
Rates can be averaged using time, distance, population or volume exposure. The denominator must match the quantity being aggregated.
For example, volume-weighted concentration answers a different question from an unweighted mean of sample concentrations.
A zero-weight observation contributes nothing to numerator or denominator. It remains in the record but does not move the weighted mean.
Explain why exclusion by zero weight is justified rather than hiding it.
Some advanced estimators use negative weights, but the result can fall outside the observed range and loses the simple “relative contribution” interpretation.
For school-level weighted averages, non-negative weights are normally the defensible default.
The ordinary mean of60,80,90 is76.67, while the weighted mean is81 because the highest score has the largest weight.
Neither number is intrinsically fairer; fairness depends on the declared assessment design.
Increasing the weight on90 moves the mean toward90. The maximum influence comes from the combination of a large weight and a value far from the current centre.
Display contribution w_i x_i, not weights alone.
Multiplying every weight by the same positive constant leaves numerator and denominator multiplied by that constant. The ratio stays unchanged.
This invariance is a strong implementation check.
If weights are dimensionless, the weighted mean has the value’s unit. If weights carry exposure units, the numerator and denominator should still simplify to the intended unit.
Write units through the formula to catch mismatches.
A missing value with a present weight cannot simply enter as zero. Remove or impute the pair under a declared rule and recompute the denominator.
Preserve the original row and reason for exclusion.
Rounded weights may sum to0.99 or1.01. Dividing by the actual sum prevents a small systematic error.
Keep higher precision internally and round the final mean appropriately.
Students can allocate ten counters across three scores to represent weights2,3,5. Expanding repeated values makes the formula concrete.
Then redistribute one counter and predict the direction of change.
Place values and weights in adjacent columns, calculate each product, use SUMPRODUCT for the numerator and SUM for the denominator.
Add checks for missing pairs, negative weights and a zero total weight.
Validate equal array lengths, finite pairs and non-zero total weight. Keep weight type in metadata.
Unit-test scores60,80,90 with weights0.2,0.3,0.5 to return81.
A weighted descriptive mean is not automatically an unbiased population estimate. Survey, inverse-variance and sampling weights require design-specific uncertainty.
Do not attach an ordinary standard error without checking the weight model.
Weights do not create information. A heavily weighted poor measurement can dominate the result.
A weighted mean also hides distribution shape, disagreement and subgroup gaps.
Weighted means appear in economics, engineering, public policy, education and AI loss functions. Analysts use them whenever contributions differ by design.
The transferable skill is tracing every weight to an auditable reason.
The Mandel–Paule guide shows a more advanced weighted consensus. The Birge-ratio guide audits whether weighted scatter agrees with stated uncertainties.
Students should label the weight type, show products and divide by the actual weight sum. A weighted result without its rule is incomplete.
Parents can ask who chose the weights and what learning purpose they serve.
Find the weighted mean of70,80,100 with weights1,2,1. Double all weights and verify invariance. Set the last weight to zero and explain the change.
Compare frequency and percentage interpretations.
### Must weights total 100%? No. Divide by their actual sum.
### Can weights be counts? Yes. Frequency weights reproduce an expanded dataset.
### Is a weighted mean always between the values? With non-negative weights and positive total weight, yes.
### Do larger weights mean better data? Only if the design explicitly defines them that way.
Read the official NIST weighted-mean reference and NIST Dataplot overview. The worked products and frequency example were independently recalculated.
Explore the Mathematics Learning Hub and How Mathematics Works.
Weighted mean turns an average into a statement about contribution. Good mathematics makes the weights as visible and testable as the values they combine.
Extended practice laboratory
Practice case 1: List values
Setup: Use the article’s weighted-mean example and change one declared input for case 1. Task: Recalculate from raw values, preserve units and state which mathematical or design condition changed.
Check: Check the relevant invariant, separate the number from its claim, and name the next diagnostic or source of evidence. Rebuild the relevant counts before interpreting the coefficient, preserve the stated design, and write one sentence separating the numerical result from the claim it can support. Then vary one defensible assumption, calculate again, and explain whether the conclusion changes because of new data, a new reference model, or a new decision purpose. Finish by naming the unit, tolerance, and evidence needed for transfer. Record one graph or table that would expose failure hidden by the summary.
Practice case 2: List weights
Setup: Use the article’s weighted-mean example and change one declared input for case 2. Task: Recalculate from raw values, preserve units and state which mathematical or design condition changed.
Check: Check the relevant invariant, separate the number from its claim, and name the next diagnostic or source of evidence. Rebuild the relevant counts before interpreting the coefficient, preserve the stated design, and write one sentence separating the numerical result from the claim it can support. Then vary one defensible assumption, calculate again, and explain whether the conclusion changes because of new data, a new reference model, or a new decision purpose. Finish by naming the unit, tolerance, and evidence needed for transfer. Record one graph or table that would expose failure hidden by the summary.
Practice case 3: Multiply value by weight
Setup: Use the article’s weighted-mean example and change one declared input for case 3. Task: Recalculate from raw values, preserve units and state which mathematical or design condition changed.
Check: Check the relevant invariant, separate the number from its claim, and name the next diagnostic or source of evidence. Rebuild the relevant counts before interpreting the coefficient, preserve the stated design, and write one sentence separating the numerical result from the claim it can support. Then vary one defensible assumption, calculate again, and explain whether the conclusion changes because of new data, a new reference model, or a new decision purpose. Finish by naming the unit, tolerance, and evidence needed for transfer. Record one graph or table that would expose failure hidden by the summary.
Practice case 4: Sum weighted values
Setup: Use the article’s weighted-mean example and change one declared input for case 4. Task: Recalculate from raw values, preserve units and state which mathematical or design condition changed.
Check: Check the relevant invariant, separate the number from its claim, and name the next diagnostic or source of evidence. Rebuild the relevant counts before interpreting the coefficient, preserve the stated design, and write one sentence separating the numerical result from the claim it can support. Then vary one defensible assumption, calculate again, and explain whether the conclusion changes because of new data, a new reference model, or a new decision purpose. Finish by naming the unit, tolerance, and evidence needed for transfer. Record one graph or table that would expose failure hidden by the summary.
Practice case 5: Sum weights
Setup: Use the article’s weighted-mean example and change one declared input for case 5. Task: Recalculate from raw values, preserve units and state which mathematical or design condition changed.
Check: Check the relevant invariant, separate the number from its claim, and name the next diagnostic or source of evidence. Rebuild the relevant counts before interpreting the coefficient, preserve the stated design, and write one sentence separating the numerical result from the claim it can support. Then vary one defensible assumption, calculate again, and explain whether the conclusion changes because of new data, a new reference model, or a new decision purpose. Finish by naming the unit, tolerance, and evidence needed for transfer. Record one graph or table that would expose failure hidden by the summary.
Practice case 6: Divide by weight sum
Setup: Use the article’s weighted-mean example and change one declared input for case 6. Task: Recalculate from raw values, preserve units and state which mathematical or design condition changed.
Check: Check the relevant invariant, separate the number from its claim, and name the next diagnostic or source of evidence. Rebuild the relevant counts before interpreting the coefficient, preserve the stated design, and write one sentence separating the numerical result from the claim it can support. Then vary one defensible assumption, calculate again, and explain whether the conclusion changes because of new data, a new reference model, or a new decision purpose. Finish by naming the unit, tolerance, and evidence needed for transfer. Record one graph or table that would expose failure hidden by the summary.
Practice case 7: Normalise weights
Setup: Use the article’s weighted-mean example and change one declared input for case 7. Task: Recalculate from raw values, preserve units and state which mathematical or design condition changed.
Check: Check the relevant invariant, separate the number from its claim, and name the next diagnostic or source of evidence. Rebuild the relevant counts before interpreting the coefficient, preserve the stated design, and write one sentence separating the numerical result from the claim it can support. Then vary one defensible assumption, calculate again, and explain whether the conclusion changes because of new data, a new reference model, or a new decision purpose. Finish by naming the unit, tolerance, and evidence needed for transfer. Record one graph or table that would expose failure hidden by the summary.
Practice case 8: Use percentages
Setup: Use the article’s weighted-mean example and change one declared input for case 8. Task: Recalculate from raw values, preserve units and state which mathematical or design condition changed.
Check: Check the relevant invariant, separate the number from its claim, and name the next diagnostic or source of evidence. Rebuild the relevant counts before interpreting the coefficient, preserve the stated design, and write one sentence separating the numerical result from the claim it can support. Then vary one defensible assumption, calculate again, and explain whether the conclusion changes because of new data, a new reference model, or a new decision purpose. Finish by naming the unit, tolerance, and evidence needed for transfer. Record one graph or table that would expose failure hidden by the summary.
Practice case 9: Use frequency weights
Setup: Use the article’s weighted-mean example and change one declared input for case 9. Task: Recalculate from raw values, preserve units and state which mathematical or design condition changed.
Check: Check the relevant invariant, separate the number from its claim, and name the next diagnostic or source of evidence. Rebuild the relevant counts before interpreting the coefficient, preserve the stated design, and write one sentence separating the numerical result from the claim it can support. Then vary one defensible assumption, calculate again, and explain whether the conclusion changes because of new data, a new reference model, or a new decision purpose. Finish by naming the unit, tolerance, and evidence needed for transfer. Record one graph or table that would expose failure hidden by the summary.
Practice case 10: Add a zero weight
Setup: Use the article’s weighted-mean example and change one declared input for case 10. Task: Recalculate from raw values, preserve units and state which mathematical or design condition changed.
Check: Check the relevant invariant, separate the number from its claim, and name the next diagnostic or source of evidence. Rebuild the relevant counts before interpreting the coefficient, preserve the stated design, and write one sentence separating the numerical result from the claim it can support. Then vary one defensible assumption, calculate again, and explain whether the conclusion changes because of new data, a new reference model, or a new decision purpose. Finish by naming the unit, tolerance, and evidence needed for transfer. Record one graph or table that would expose failure hidden by the summary.
Practice case 11: Double all weights
Setup: Use the article’s weighted-mean example and change one declared input for case 11. Task: Recalculate from raw values, preserve units and state which mathematical or design condition changed.
Check: Check the relevant invariant, separate the number from its claim, and name the next diagnostic or source of evidence. Rebuild the relevant counts before interpreting the coefficient, preserve the stated design, and write one sentence separating the numerical result from the claim it can support. Then vary one defensible assumption, calculate again, and explain whether the conclusion changes because of new data, a new reference model, or a new decision purpose. Finish by naming the unit, tolerance, and evidence needed for transfer. Record one graph or table that would expose failure hidden by the summary.
Practice case 12: Change one high weight
Setup: Use the article’s weighted-mean example and change one declared input for case 12. Task: Recalculate from raw values, preserve units and state which mathematical or design condition changed.
Check: Check the relevant invariant, separate the number from its claim, and name the next diagnostic or source of evidence. Rebuild the relevant counts before interpreting the coefficient, preserve the stated design, and write one sentence separating the numerical result from the claim it can support. Then vary one defensible assumption, calculate again, and explain whether the conclusion changes because of new data, a new reference model, or a new decision purpose. Finish by naming the unit, tolerance, and evidence needed for transfer. Record one graph or table that would expose failure hidden by the summary.
Practice case 13: Compare arithmetic mean
Setup: Use the article’s weighted-mean example and change one declared input for case 13. Task: Recalculate from raw values, preserve units and state which mathematical or design condition changed.
Check: Check the relevant invariant, separate the number from its claim, and name the next diagnostic or source of evidence. Rebuild the relevant counts before interpreting the coefficient, preserve the stated design, and write one sentence separating the numerical result from the claim it can support. Then vary one defensible assumption, calculate again, and explain whether the conclusion changes because of new data, a new reference model, or a new decision purpose. Finish by naming the unit, tolerance, and evidence needed for transfer. Record one graph or table that would expose failure hidden by the summary.
Practice case 14: Check units
Setup: Use the article’s weighted-mean example and change one declared input for case 14. Task: Recalculate from raw values, preserve units and state which mathematical or design condition changed.
Check: Check the relevant invariant, separate the number from its claim, and name the next diagnostic or source of evidence. Rebuild the relevant counts before interpreting the coefficient, preserve the stated design, and write one sentence separating the numerical result from the claim it can support. Then vary one defensible assumption, calculate again, and explain whether the conclusion changes because of new data, a new reference model, or a new decision purpose. Finish by naming the unit, tolerance, and evidence needed for transfer. Record one graph or table that would expose failure hidden by the summary.
Practice case 15: Handle missing pairs
Setup: Use the article’s weighted-mean example and change one declared input for case 15. Task: Recalculate from raw values, preserve units and state which mathematical or design condition changed.
Check: Check the relevant invariant, separate the number from its claim, and name the next diagnostic or source of evidence. Rebuild the relevant counts before interpreting the coefficient, preserve the stated design, and write one sentence separating the numerical result from the claim it can support. Then vary one defensible assumption, calculate again, and explain whether the conclusion changes because of new data, a new reference model, or a new decision purpose. Finish by naming the unit, tolerance, and evidence needed for transfer. Record one graph or table that would expose failure hidden by the summary.
Practice case 16: Reject negative weights
Setup: Use the article’s weighted-mean example and change one declared input for case 16. Task: Recalculate from raw values, preserve units and state which mathematical or design condition changed.
Check: Check the relevant invariant, separate the number from its claim, and name the next diagnostic or source of evidence. Rebuild the relevant counts before interpreting the coefficient, preserve the stated design, and write one sentence separating the numerical result from the claim it can support. Then vary one defensible assumption, calculate again, and explain whether the conclusion changes because of new data, a new reference model, or a new decision purpose. Finish by naming the unit, tolerance, and evidence needed for transfer. Record one graph or table that would expose failure hidden by the summary.
Practice case 17: Preserve categories
Setup: Use the article’s weighted-mean example and change one declared input for case 17. Task: Recalculate from raw values, preserve units and state which mathematical or design condition changed.
Check: Check the relevant invariant, separate the number from its claim, and name the next diagnostic or source of evidence. Rebuild the relevant counts before interpreting the coefficient, preserve the stated design, and write one sentence separating the numerical result from the claim it can support. Then vary one defensible assumption, calculate again, and explain whether the conclusion changes because of new data, a new reference model, or a new decision purpose. Finish by naming the unit, tolerance, and evidence needed for transfer. Record one graph or table that would expose failure hidden by the summary.
Practice case 18: Audit rounding
Setup: Use the article’s weighted-mean example and change one declared input for case 18. Task: Recalculate from raw values, preserve units and state which mathematical or design condition changed.
Check: Check the relevant invariant, separate the number from its claim, and name the next diagnostic or source of evidence. Rebuild the relevant counts before interpreting the coefficient, preserve the stated design, and write one sentence separating the numerical result from the claim it can support. Then vary one defensible assumption, calculate again, and explain whether the conclusion changes because of new data, a new reference model, or a new decision purpose. Finish by naming the unit, tolerance, and evidence needed for transfer. Record one graph or table that would expose failure hidden by the summary.
Practice case 19: Plot contributions
Setup: Use the article’s weighted-mean example and change one declared input for case 19. Task: Recalculate from raw values, preserve units and state which mathematical or design condition changed.
Check: Check the relevant invariant, separate the number from its claim, and name the next diagnostic or source of evidence. Rebuild the relevant counts before interpreting the coefficient, preserve the stated design, and write one sentence separating the numerical result from the claim it can support. Then vary one defensible assumption, calculate again, and explain whether the conclusion changes because of new data, a new reference model, or a new decision purpose. Finish by naming the unit, tolerance, and evidence needed for transfer. Record one graph or table that would expose failure hidden by the summary.
Practice case 20: Compare two weighting schemes
Setup: Use the article’s weighted-mean example and change one declared input for case 20. Task: Recalculate from raw values, preserve units and state which mathematical or design condition changed.
Check: Check the relevant invariant, separate the number from its claim, and name the next diagnostic or source of evidence. Rebuild the relevant counts before interpreting the coefficient, preserve the stated design, and write one sentence separating the numerical result from the claim it can support. Then vary one defensible assumption, calculate again, and explain whether the conclusion changes because of new data, a new reference model, or a new decision purpose. Finish by naming the unit, tolerance, and evidence needed for transfer. Record one graph or table that would expose failure hidden by the summary.
Practice case 21: Check fairness claim
Setup: Use the article’s weighted-mean example and change one declared input for case 21. Task: Recalculate from raw values, preserve units and state which mathematical or design condition changed.
Check: Check the relevant invariant, separate the number from its claim, and name the next diagnostic or source of evidence. Rebuild the relevant counts before interpreting the coefficient, preserve the stated design, and write one sentence separating the numerical result from the claim it can support. Then vary one defensible assumption, calculate again, and explain whether the conclusion changes because of new data, a new reference model, or a new decision purpose. Finish by naming the unit, tolerance, and evidence needed for transfer. Record one graph or table that would expose failure hidden by the summary.
Practice case 22: Estimate uncertainty
Setup: Use the article’s weighted-mean example and change one declared input for case 22. Task: Recalculate from raw values, preserve units and state which mathematical or design condition changed.
Check: Check the relevant invariant, separate the number from its claim, and name the next diagnostic or source of evidence. Rebuild the relevant counts before interpreting the coefficient, preserve the stated design, and write one sentence separating the numerical result from the claim it can support. Then vary one defensible assumption, calculate again, and explain whether the conclusion changes because of new data, a new reference model, or a new decision purpose. Finish by naming the unit, tolerance, and evidence needed for transfer. Record one graph or table that would expose failure hidden by the summary.
Practice case 23: Avoid causal language
Setup: Use the article’s weighted-mean example and change one declared input for case 23. Task: Recalculate from raw values, preserve units and state which mathematical or design condition changed.
Check: Check the relevant invariant, separate the number from its claim, and name the next diagnostic or source of evidence. Rebuild the relevant counts before interpreting the coefficient, preserve the stated design, and write one sentence separating the numerical result from the claim it can support. Then vary one defensible assumption, calculate again, and explain whether the conclusion changes because of new data, a new reference model, or a new decision purpose. Finish by naming the unit, tolerance, and evidence needed for transfer. Record one graph or table that would expose failure hidden by the summary.
Practice case 24: Write a weighted-average conclusion
Setup: Use the article’s weighted-mean example and change one declared input for case 24. Task: Recalculate from raw values, preserve units and state which mathematical or design condition changed.
Check: Check the relevant invariant, separate the number from its claim, and name the next diagnostic or source of evidence. Rebuild the relevant counts before interpreting the coefficient, preserve the stated design, and write one sentence separating the numerical result from the claim it can support. Then vary one defensible assumption, calculate again, and explain whether the conclusion changes because of new data, a new reference model, or a new decision purpose. Finish by naming the unit, tolerance, and evidence needed for transfer. Record one graph or table that would expose failure hidden by the summary.