PSLE Science Reality Lab Vol No.274 | “Water Quality Index = 85” — Is the Water 85% Clean?

Series ID: PSLE-SCI-REALITY-0274

Wait, What? A Score of 85 Can Mean “Good” in One Water Index and Something Very Different in Another

A student opens a river dashboard. Beside River A is a large number: Water Quality Index: 85. The obvious-looking interpretation is, “The water is 85% clean.” It feels reasonable because 85 looks like a percentage, and school marks have trained us to read 85 out of 100 as a strong result.

But a water-quality index is not automatically a percentage of cleanliness. It is a constructed score. Scientists and agencies choose which water properties matter for a stated purpose, compare observations with guidelines or other reference rules, and combine the evidence according to a particular index method. The number is useful precisely because it compresses many measurements into one communication object.

There is an even bigger surprise. The Canadian Council of Ministers of the Environment Water Quality Index runs from 0 for worst to 100 for best. An older British Columbia index used a 0-to-100 direction in which 0 represented the best condition and 100 the poorest. So even the direction of the scale cannot be guessed from the number alone.

The Reality Lab habit is therefore simple and powerful: before interpreting an index, reconstruct what went into it, what direction the scale runs, and what question it was designed to answer.

Quick Answer

A Water Quality Index, or WQI, is a summary score built from selected water-quality evidence under a defined method.
A score of 85 does not mean that 85% of the water is clean.
It does not mean that every measured property scored 85.
It does not mean that every possible contaminant was tested.
It does not necessarily tell you whether water is safe to drink, swim in, irrigate with or support aquatic life unless the index is specifically designed for that use.
Different indices can choose different parameters, guidelines, formulas, categories and scale directions.
The scientific job is to identify the index, its purpose, its inputs, its reference rules, its time and place coverage, and any missing information.
A summary score can be useful without being a literal physical quantity.

The Exact Learner Job This Reality Lab Owns

This volume owns one evidence-transfer job: how to evaluate a real-world water-quality index shown on a report, dashboard, infographic or news graphic without mistaking the index score for a percentage clean, a direct concentration, a universal safety rating or a complete description of the water.

It does not own water chemistry, pH, turbidity, dissolved oxygen, environmental sampling, graph reading, measurement uncertainty or the mathematics of index construction. Those remain with existing Science owners. It also does not replace Reality Lab Vol No.028, which owns the broader problem of a composite overall score hiding its parts. This volume applies that evidence habit to a particularly important scientific communication object: a water-quality score whose meaning depends on purpose, parameters and reference rules.

Why This Is a PSLE Science Evidence Job

The current 2026 PSLE Science assessment framework assesses Knowledge with Understanding together with Application of Knowledge and Scientific Inquiry. That inquiry includes interpreting and analysing information, evaluating observations, information and methods, and communicating explanations and reasoning. The 2023 Primary Science syllabus also encourages evidence-based thinking, healthy scepticism, attention to assumptions and uncertainty, and understanding that Science is communicated in different forms and media.

A WQI is exactly the kind of object that rewards those habits. It looks like one easy number. Yet the number sits at the end of a chain: samples were collected, measurements were made, guidelines or reference conditions were selected, rules transformed those measurements, and a score was communicated. A strong learner learns to travel backward through that chain before making a claim.

Rebuild the Evidence Object: The Fictional River Card

Imagine a fictional environmental dashboard for Bluebird River. It shows only this:

Station	Water Quality Index	Colour
Bluebird River	85	Green

A student writes, “The river is 85% clean.” What is missing?

Which WQI method produced 85?
Does a higher number mean better or worse quality?
Which properties were measured?
Which guidelines or reference values were used?
What intended use was being assessed?
How many samples were collected, where and when?
Were some parameters missing?
Does the score describe one monitoring station, a river reach, a time period or a larger region?

Without those answers, 85 is a label with incomplete meaning. The number is not useless; it is simply method-dependent evidence.

Observed, Compared, Calculated, Communicated, Inferred

Layer	Example	What it can support
Observed	pH, dissolved oxygen, turbidity, temperature or chemical concentrations measured in samples	Evidence about those measured properties under the sampling conditions
Compared	Measurements checked against guidelines or reference values	Evidence about whether stated reference conditions were met
Calculated	Several results transformed and combined into WQI = 85	A summary under that index method
Communicated	Green colour and “good” category	A simplified public interpretation defined by the system
Unsupported inference	“The water is 85% clean and safe for every use”	Too broad unless the method and evidence specifically support it

The mistake occurs when the final communication layer is silently converted into a different physical claim.

A WQI Is a Summary, Not a Substance in the Water

If a laboratory reports nitrate as 5 mg/L, the number has a physical quantity and unit. If a meter reports pH 7.2, the number belongs to a defined measurement scale. A WQI score is different. It is generally a derived index made by combining information according to rules.

The Canadian CCME WQI, for example, summarizes complex water-quality data using three broad ingredients: the scope of guideline failures, their frequency, and their amplitude. It then produces a score between 0 and 100, with 100 representing the best quality under that method. The specific parameters, guidelines and time period can vary with local conditions and purpose.

That is an important scientific clue. The index is not “the fraction of water molecules that are clean.” It is a compact statement about how a chosen set of observations behaved relative to chosen reference rules.

The Same Number Can Mean Different Things Under Different Index Systems

Suppose Dashboard A uses an index where 100 is best. Dashboard B uses a different system where 100 is worst. Both show 85.

It would be scientifically invalid to say the sites have the same water quality merely because the numerals match. First align the definitions. In science, numbers do not travel safely without their quantity, unit or scoring rule.

This also means a student should not memorise “85 = good” as a universal rule. The correct habit is: read the legend and method before reading the rank.

Purpose Check: Good for What?

Water quality is not one universal property. The U.S. Geological Survey describes water quality in relation to suitability for particular uses and characteristics. An index designed to evaluate water for aquatic life can emphasise different evidence from one designed for irrigation or public water supply.

Therefore, the phrase “good water quality” needs a scientific receiver: good for what purpose under which criteria?

Do not turn this into a safety shortcut. A learner should never use a generic dashboard score to decide whether untreated environmental water is personally safe to drink. That is a different claim requiring appropriate authoritative testing and public-health guidance.

Input Check: Which Water Properties Went Into the Score?

Imagine two fictional WQI systems.

Index Alpha	Index Beta
pH	pH
dissolved oxygen	turbidity
temperature	nitrate
nitrate	phosphate
turbidity	conductivity

Even if both systems return 85, they did not summarize identical evidence. A value missing from the input set cannot be magically known because the final index looks comprehensive.

The key question is not “How many variables are there?” but “Are these variables relevant to the stated use, and were they measured well enough to support the claim?”

One Bad Component Can Hide Inside a Respectable-Looking Score

Suppose a fictional index averages five transformed component scores:

Component	Score
A	100
B	100
C	95
D	90
E	40

The overall score might still look quite strong. Yet component E is much poorer than the others. Real WQI systems do not all use simple averaging, but the lesson survives: compression can hide the distribution of the ingredients.

When a claim matters, inspect the component data rather than treating the headline score as the whole evidence packet.

Sampling Check: A Score Can Only Summarize the Evidence That Was Collected

A WQI can look geographically large on a map even when it was calculated from a limited monitoring network. Ask where samples came from, how often they were taken, whether wet and dry seasons were represented, and whether the monitored location can stand for the area shown.

One grab sample is a snapshot, not the whole river through the whole day. That job is already owned by Reality Lab Vol No.246. Here, the transfer is to remember that a polished index cannot expand the spatial or temporal reach of its inputs.

Representation Check: The Green Circle Is Not Another Measurement

Dashboards often convert numerical scores into colours: green, yellow, orange or red. Colour helps readers scan quickly, but it adds another decision layer. Someone selected category boundaries. A score of 79 and 80 may receive different colours even though the numerical difference is small.

Therefore, ask whether the dramatic visual boundary is a physical discontinuity in the river or a communication threshold imposed by the index. A colour change can be useful without meaning nature itself changed suddenly at that exact number.

Worked Case 1: “River A Scored 90, So It Is 90% Pure”

Tempting reasoning: 90 out of 100 means 90% pure.

Better reasoning: Purity is a different scientific claim. A WQI score is calculated from a defined set of measurements and rules. Check the index method and parameter data. Do not convert the score into a chemical composition percentage.

Worked Case 2: Two Rivers Both Score 82

River P scores 82 using Index Alpha. River Q scores 82 using Index Beta. Can we say the water quality is equal?

No. First determine whether the methods use the same scale direction, parameters, guidelines, aggregation rules, intended use and sampling window. Matching numerals are not enough to establish scientific comparability.

Worked Case 3: The Index Improved, but One Parameter Worsened

A river’s index rises from 78 to 84. However, nitrate concentration increases while several other measurements improve. Is it correct to say “every aspect of the water improved”?

No. The overall score improved under the index, but the component data show that at least one measured property moved in the opposite direction. The index summarizes; it does not erase component-level evidence.

Worked Case 4: The Score Fell After More Parameters Were Added

Year 1 used five parameters. Year 2 used ten, including several that sometimes exceeded their guidelines. The WQI fell. Does that prove the river physically became worse?

Not by itself. The measurement system changed. A fair trend analysis asks whether the same input set and rules were used or whether the earlier score can be recalculated on a comparable basis.

This routes to the broader measurement-system lesson in Reality Lab Vol No.048.

Worked Case 5: pH Passed but Turbidity Rose

A poster says, “The river is fine because pH is normal.” That is a one-parameter argument for a multi-property claim. pH is important, but it is not a universal detector for every water-quality issue. Likewise, turbidity is an optical measurement, not a direct particle count or complete safety verdict.

For the pH scale, route to Reality Lab Vol No.207. For turbidity, route to Reality Lab Vol No.171. Reality Lab 274 does not re-teach either measurement; it asks how multiple measurements become one public score.

Alternative Explanations for a Changing WQI

If a WQI changes from one period to the next, several explanations may fit:

the underlying water conditions truly changed;
rainfall, flow or seasonal conditions changed;
the sampling locations changed;
the number of samples changed;
different parameters were included;
guidelines or reference values changed;
laboratory methods changed;
a few large exceedances strongly affected the index;
missing data altered which inputs were available.

The index trend is an observation about the calculated score. It does not identify the cause by itself.

Evidence That Strengthens a WQI Claim

The exact index method is named.
The scale direction and categories are explained.
The intended water use is stated.
The included parameters and guideline values are available.
Sampling locations, frequency and time period are clear.
The same method is used for comparisons through time or across sites.
Missing data and exclusions are reported.
Component measurements are available alongside the summary score.
The conclusion stays within the index’s design and sampling scope.

Evidence That Weakens an Over-Broad WQI Claim

The report says only “water quality score” without naming the method.
A score is described as a percentage clean.
The index is used to claim safety for a purpose it was not designed to assess.
Different index systems are compared by raw number.
The colour category is shown without the scale or legend.
A single station is presented as the whole river without justification.
The parameter list changed but the trend is presented as directly comparable.
A high overall score is used to imply every component is good.
Unmeasured substances are treated as though the index proved they were absent.

Do Not Overcorrect: A Water Quality Index Is Not a Trick

After learning how much information is compressed into an index, a student may say, “Then WQI scores are meaningless.” That is also incorrect.

Indices solve a genuine communication problem. Large monitoring programmes can produce many measurements over many dates. A carefully defined index can summarize whether important guidelines are frequently or substantially exceeded and help readers compare patterns. The CCME explicitly describes its WQI as a convenient way to summarize complex water-quality data and communicate it to a general audience.

The scientific skill is calibrated trust: use the score for the job it was designed to do, and return to the components when the claim becomes more specific.

How Far Can the Conclusion Travel?

Suppose a monitoring programme states that Station A had a CCME WQI of 85 for a defined period using parameters and guidelines chosen for protection of aquatic life. A careful conclusion is that under that method, parameter set, guideline framework, sampling design and period, the station received a high index score for that stated water-quality purpose.

The number alone does not establish that:

85% of the water is clean;
the water is 85% pure;
every parameter scored 85;
all possible contaminants were tested;
the entire river has the same condition;
the result remains the same every day;
the water is safe for every possible use;
another WQI score of 85 is directly comparable.

Tempting but Invalid Reasoning

“85 means 85% clean.” An index score is not automatically a percent composition.
“Higher is always better.” Check the direction of the named index.
“Same score means same water.” Different component patterns can lead to similar summary scores.
“The score is green, so every parameter passed.” Inspect the component data and index rules.
“The river score is 90, so untreated water is safe to drink.” A general index is not a substitute for authoritative drinking-water testing and guidance.
“The score fell, so pollution definitely increased.” Method, sampling and seasonal changes are alternative explanations.
“The index includes five variables, so everything important was measured.” Unmeasured properties remain unmeasured.

PSLE-Style Transfer Case: The Lake Poster

A fictional lake poster gives these results:

Year	WQI	Parameters used
2025	82	pH, turbidity, dissolved oxygen, nitrate, phosphate
2026	76	pH, turbidity, dissolved oxygen, nitrate, phosphate, conductivity, temperature

Question 1: Can we immediately conclude the lake became worse?

Answer: No. The input set changed, so the two scores may not be directly comparable. We need the index method and preferably a recalculation using a consistent set of parameters.

Question 2: Does WQI 82 mean 82% of the lake was clean?

Answer: No. The WQI is a derived score, not a fraction of lake water classified as clean.

Question 3: What would strengthen a claim that water quality genuinely changed?

Answer: Comparable sampling, the same WQI method and parameter set, component-level measurements, repeated observations over appropriate times, and an explanation showing which properties changed.

Explained Practice

1. A dashboard says WQI 92. Can you call the water 92% pure? No. Purity is not what the index score directly represents.

2. Two sites use the same CCME WQI, same parameters and same period. Is comparing their scores more defensible? Yes. Important parts of the measurement system are aligned, though sampling quality still matters.

3. A high WQI is based on three samples from one location. Can it automatically describe 30 km of river? No. Spatial generalisation requires representative evidence.

4. A low component value is hidden by strong values elsewhere. Should the component be ignored? No. A summary score should not erase an important component result.

5. An index changes colour at 80. Does nature suddenly change at exactly 80? Not necessarily. The category boundary is a communication rule.

6. One index runs 0 worst to 100 best; another runs 0 best to 100 worst. Can raw values be compared? No. The scale definitions differ.

7. A river’s pH is within a guideline. Does that prove every water-quality property is acceptable? No. One parameter cannot stand in for unmeasured properties.

8. More sampling finds occasional guideline exceedances and the WQI falls. Could improved observation partly explain the change? Yes. Increased evidence can reveal events that sparse sampling missed.

9. Is it reasonable to use a WQI as a public summary when the method is transparent? Yes. Compression is useful when its boundaries are preserved.

10. What is the first question when you see “WQI = 85”? Which WQI, for what purpose, using which evidence?

Delayed Independent Return

Tomorrow, imagine a dashboard that says only “Lake Quality Score: 88” with a green icon. Write down five things you would need before interpreting it. Then add one sentence that is careful enough to remain true even before those details arrive.

A strong answer might say: “The dashboard reports a high-looking score, but its scientific meaning depends on the named index, scale direction, parameters, reference rules and sampling design.” That sentence separates observation from inference without rejecting the evidence.

Useful eduKateSengkang Routes

Parent and Tutor Teaching Guide: Build a Score, Then Break It Open

Create five fictional water measurements and give each a simple 0-to-100 classroom score. Let the learner calculate an average. Then ask what information disappeared when five values became one number. The learner should notice that the final score no longer shows which component was highest, which was lowest, or whether one result was unusual.

Next, change the rules. Tell the learner that low numbers are now better. Keep the same raw observations but change the index transformation. This demonstrates that the meaning of an index comes from its definition, not from the visual familiarity of a 0-to-100 scale.

Finally, remove one measurement from the second year. Ask whether the two years remain perfectly comparable. This creates a direct bridge from a simple classroom activity to real monitoring systems without teaching specialist environmental chemistry.

Authoritative Sources

The CCME manual is particularly useful because it states the scientific communication job clearly: the WQI summarizes complex water-quality data, while its parameters, guidelines and time period can vary with purpose and local conditions. That is exactly why a WQI number must stay attached to its method.

The Quiet Rule to Keep

When you see “Water Quality Index = 85,” do not ask first whether 85 is high. Ask what 85 means here.

Find the index. Find the purpose. Find the parameters. Find the reference rules. Find the samples. Then let the score say exactly what it was built to say—no less, and no more.

Small Group Tutorials