How to Tell Whether a PSLE Science Measurement Is the Outcome or a Check on a Controlled Condition

Wait, What? Not Every Number in an Investigation Is the Result

A PSLE Science method records several numbers:

100 mL of water in each cup;
80°C starting temperature in each cup;
ten minutes of cooling;
62°C final temperature in Cup P;
55°C final temperature in Cup Q.

A learner sees five numerical quantities and says, “They are all measured variables.”

But the numbers do different scientific jobs.

The water volume may be measured to make sure both cups start comparably. The starting temperatures may be checked for the same reason. The ten-minute duration defines the comparison interval. The final temperature—or the temperature change calculated from start to finish—may be the outcome that answers the investigation question.

A measurement is not automatically the outcome just because it produces a number. The scientific question decides which measurement is the response you are trying to explain and which measurements are there to check that the comparison stayed fair.

Quick Answer

To decide whether a PSLE Science measurement is the main outcome or a check on a controlled condition, start from the investigation question. Ask: Which quantity changes in response to the factor being tested? That is the outcome measurement. Then ask: Which other measured quantities are checked so the set-ups remain comparable? Those are control-check measurements.

Use this route:

STATE THE SCIENTIFIC QUESTION → IDENTIFY THE DELIBERATELY CHANGED FACTOR → IDENTIFY THE QUANTITY THAT ANSWERS THE QUESTION → LABEL THAT AS THE OUTCOME → IDENTIFY OTHER MEASUREMENTS USED TO VERIFY COMPARABILITY → KEEP THEIR ROLES SEPARATE → READ RESULTS → EXPLAIN THE OUTCOME → CHECK THAT CONTROL-CHECK VALUES DID NOT BECOME A SECOND TESTED FACTOR.

The Exact PSLE Science Learning Job This Guide Owns

This guide owns one Primary 5/6 learner job: distinguishing the measurement that directly answers an investigation question from measurements made only to verify, monitor or preserve controlled conditions.

It does not replace How to Decode Variables and Fair Tests, which owns the broad changed / measured / controlled-variable structure. It does not replace How to Decide What to Measure, which owns choosing a suitable outcome. And it does not replace the set-condition versus experienced-condition guide.

This page owns the boundary that becomes difficult once a method contains several numerical measurements: which measurement is the answer-bearing response, and which one is only checking the method?

Why This Matters in the Current PSLE Science Frame

For examination from 2026, PSLE Science assesses attainment in the 2023 Primary Science syllabus. SEAB’s assessment objectives include applying scientific inquiry, interpreting and analysing information, evaluating observations, information and methods, and communicating explanations and reasoning.

Those skills require learners to understand the role of evidence inside an investigation. A number in a method can describe the tested condition, the response, a baseline, a timing rule or a control check. Reading every number as the same kind of evidence makes fair-test reasoning unstable.

This guide teaches a reasoning distinction. It does not invent an official marking label that every school must use.

First Distinction: Measured Does Not Mean “Measured Outcome”

Quantity	Possible role	Question it answers
Final temperature	Outcome measurement	What happened to the system?
Starting temperature	Baseline or control check	Did the set-ups begin comparably?
Water volume	Controlled-condition check	Was the amount of water kept comparable?
Distance from lamp	Tested condition or controlled condition	Depends on the investigation question.
Time	Tested variable, controlled duration or measurement schedule	Depends on the method.

The same physical quantity can play different roles in different investigations. Role comes from the question, not from the instrument.

The Question-Ownership Test

Write the investigation as a relationship:

How does ______ affect ______?

The first blank is the deliberately changed factor. The second blank is the outcome that must be observed or measured.

Every other relevant measurement has to justify a different job: baseline, control, verification, timing, specimen matching, calibration or another method check.

Worked Example 1 — Wrapping Material and Cooling

Original practice question: How does wrapping material affect how much warm water cools over ten minutes?

Method:

Use identical cups.
Place 100 mL of water in each cup.
Check that both begin at 80°C.
Wrap one cup with Material P and one with Material Q.
Measure temperature after ten minutes.

Role map:

Information	Role
Material P vs Q	Deliberately changed condition
100 mL water	Controlled condition; volume may be measured to verify comparability
80°C starting temperature	Baseline/control check
10 min	Controlled observation duration
Final temperature or temperature decrease	Outcome evidence answering the cooling question

If a learner concludes, “Material P is better because both cups started at 80°C,” the control check has been mistaken for the outcome. Equal starting temperature helps make the comparison interpretable; it does not tell which cup cooled more.

Worked Example 2 — Ramp Height and Travel Distance

Original practice question: How does the height of a ramp affect the distance travelled by the same toy car on the same surface?

Measurements might include:

ramp height;
car mass;
starting position;
distance travelled.

Ramp height is the deliberately changed quantity. Distance travelled is the outcome. Car mass might be measured beforehand to verify that the same or comparable car is used, but it is not the outcome unless the investigation question is changed to ask about mass.

The same number can therefore be important without belonging in the final relationship statement.

Worked Example 3 — Plant Growth With Temperature Monitoring

Two groups of similar seedlings receive different stated light conditions. The learner measures plant height each day and also records room temperature to check that both groups experience comparable temperatures.

Plant-height change is the outcome. Room temperature is a control-check measurement.

If room temperature differs greatly between the groups, that check has detected a method problem. The temperature record becomes important evidence about the fairness of the comparison—but it still does not become the plant-growth outcome.

Worked Example 4 — One Instrument, Two Different Roles

A thermometer can serve different roles in the same investigation.

At the start, it may check that all set-ups begin at the same temperature.
During the investigation, it may monitor a condition intended to stay stable.
At the end, it may measure the outcome if temperature change is what the investigation asks about.

Do not label the instrument as “the outcome instrument” forever. The scientific role belongs to each measurement event.

Worked Example 5 — Same Quantity, Different Investigation

Investigation A asks how material type affects temperature decrease. Temperature is the outcome.

Investigation B asks how temperature affects the time taken for a process. Temperature is now the deliberately changed condition; time is the outcome.

Investigation C asks how light affects growth while temperature is kept comparable. Temperature is now a controlled-condition check.

Same quantity. Three different scientific roles. This is why role must be derived from the question.

Worked Example 6 — A Control Check That Reveals Drift

A learner intends room temperature to stay similar throughout repeated trials. Temperature is checked before each trial:

Trial	Room temperature / °C	Main outcome / units
1	24	8
2	25	9
3	28	12
4	31	14

The temperature column is not the intended outcome. But it reveals that a controlled condition is drifting strongly across trials. This can weaken the interpretation of the main outcome.

A control-check measurement can therefore become scientifically decisive without changing its role into the dependent outcome.

Control Check vs Baseline: Related, but Not Identical

A baseline records the starting state so later change can be calculated or interpreted.

A control check verifies that a relevant condition intended to stay comparable actually remained comparable.

One starting measurement can do both jobs in some methods. For example, starting temperature may be needed to calculate temperature decrease and also to confirm that two set-ups began at comparable temperatures.

Scientific roles can overlap. The learner should identify every job the measurement actually performs rather than force one label when two are valid.

Outcome Measurement vs Indicator Measurement

Sometimes the outcome cannot be observed directly and an indicator is measured instead.

For example, a supplied question may tell you that a colour change indicates the presence or amount of something else. Then the measured indicator is the evidence used to infer the outcome.

Use How to Use Indirect Evidence when that proxy relationship is the difficult part.

Outcome Measurement vs Process Check

A method may record a quantity to confirm that a procedure occurred correctly.

Example: the scientific question asks about growth after seven days, but the learner records water volume daily to confirm equal watering. Daily water volume is a process/control check. The final growth measure remains the outcome.

Do not let the number of control checks overwhelm the one measurement that answers the scientific question.

Where Should Each Measurement Appear in a Table?

A results table should make roles clear.

If a control check is important to interpreting the trial, it may deserve its own column or a method note. But the outcome column should remain identifiable as the response to the deliberately changed factor.

Do not build a table where every numerical condition is mixed into one row without labels. Use quantity names and units so the relationship remains visible.

The “If This Changed, Would It Change the Question?” Test

Ask about a measurement:

If this quantity changed across the intended test conditions, would that be the response the question is trying to study—or would it introduce an unwanted extra difference?

If it is the response being studied, it is likely the outcome.

If its change would make the comparison unfair, it is likely a controlled condition that may be monitored or checked.

Do Not Assume a Controlled Condition Is Perfectly Constant Because It Was Set Once

A method can set two rooms to the same temperature at the start, but temperature may drift later. A control-check measurement can reveal whether comparability survived.

This is where control measurements become powerful. They do not answer the main question directly; they tell you whether the main answer is trustworthy.

Do Not Turn the Control Check Into a Second Cause Without Evidence

If room temperature differs slightly but the outcome differs greatly, do not immediately claim the temperature difference caused the result. First judge whether the difference is scientifically relevant and large enough to matter in the context.

A control check can reveal a possible confounder. It does not automatically prove how much that confounder contributed.

What If the Control Check Fails?

If an important controlled condition is not comparable:

identify which condition failed;
explain how it could also affect the outcome;
decide whether the trial is still interpretable;
repair or standardise the method where possible;
repeat the relevant comparison if appropriate;
narrow the conclusion if the evidence remains confounded.

Do not simply delete the control-check values because they make the experiment messy. They are evidence about method quality.

What If the Outcome Measurement Also Needs a Baseline?

Suppose the outcome is “change in mass”. You need a starting mass and a final mass. Neither value alone is the outcome of interest; the change derived from them answers the question.

Use the calculated-value guide for derived outcomes.

The important distinction remains: those measurements belong to the answer-bearing outcome chain, while a separate room-temperature check may belong to method control.

Earliest-Weak-Link Diagnosis

Failure signature	Earliest weak link	Repair
Calls every recorded number the dependent variable.	Measurement role not derived from question.	Write “How does X affect Y?” first.
Uses equal starting temperature as proof one wrapping is better.	Control check mistaken for outcome.	Identify what changed after the test.
Ignores a drifting control-check measurement.	Method-quality evidence dismissed.	Ask whether drift could affect the outcome.
Calls a monitored control the deliberately changed variable.	Recorded does not equal manipulated.	Check which quantity the method intentionally varies.
Forgets baseline values needed for a change outcome.	Outcome evidence chain incomplete.	Identify which measurements are needed to derive the response.
Changes a control-check quantity and still claims the original question was tested.	Control became unintended second variable.	Restore comparability or narrow conclusion.

Misconception Repair — “If I Used an Instrument, It Must Be the Outcome”

No. Instruments can measure inputs, controls, baselines, process checks and outcomes. The question decides the role.

Misconception Repair — “Controlled Means I Never Need to Measure It”

Sometimes a controlled condition can be set reliably without repeated measurement. Sometimes it can drift and monitoring is useful. The method and scientific context decide. Do not invent unnecessary measurements, but do not assume important controls stayed equal merely because the plan said they should.

Misconception Repair — “The Outcome Is Always the Last Number Recorded”

No. The outcome could be a change calculated from start and end values, a rate, a count, a qualitative observation or another measure directly aligned with the scientific question.

Misconception Repair — “A Failed Control Check Makes Every Result Useless”

Not automatically. It weakens the ability to isolate the intended factor. How serious the problem is depends on the size and scientific relevance of the uncontrolled difference. State what the evidence can still support.

The Measurement-Role Protocol

State the scientific question.
Name the deliberately changed factor.
Name the outcome that would answer the question.
List every measurement in the method.
For each measurement ask: outcome, baseline, control check, timing, specimen matching or other method check?
Keep units attached.
Check whether any controlled quantity actually drifted.
Use only the outcome evidence to state the main relationship.
Use control-check evidence to judge fairness and method quality.
Build the causal explanation only after roles are stable.

Original Practice Set

Practice A — Cooling

Question: How does lid type affect cooling over 15 minutes? Measurements: starting temperature, water volume, final temperature.

Receipt: lid type = changed condition; final temperature/change = outcome; starting temperature and water volume = baseline/control checks.

Practice B — Light and Growth

Question: How does light condition affect height increase? Measurements: light level, initial height, final height, room temperature.

Receipt: light level = changed condition; height increase derived from initial/final heights = outcome; room temperature = control-check measurement if intended to stay comparable.

Practice C — Reverse the Question

New question: How does temperature affect plant height increase?

Temperature changes role. It is now the tested factor, not merely a control check. The outcome remains height increase.

This is the strongest test of understanding: can the learner reassign roles when the scientific question changes?

Retrieval and Practice Sequence

Start with one changed factor, one outcome and one control check.
Add a baseline needed for the outcome.
Add two control checks.
Use the same instrument for a baseline and outcome measurement.
Make one control check drift and ask how interpretation changes.
Reverse the scientific question so a former control becomes the tested factor.
Use a table containing outcome and control-check columns.
Return later with a different Science theme and no role labels.

Unfamiliar Transfer Challenge

A fictional device is tested under input levels P and Q. The method records:

input level;
device starting temperature;
room humidity;
output distance after 30 seconds.

The scientific question is: “How does input level affect output distance?”

You do not need to know what the device is.

Input level = changed factor.
Output distance = outcome.
Starting temperature and room humidity can be control-check measurements if the method intends them to stay comparable and they could affect the response.
30 seconds = comparison duration.

The learner operation transfers because it begins from the question, not the topic name.

Delayed Independent Return Test

Three to five days later, take a fresh investigation containing at least four measured quantities. Without notes:

write the scientific question;
name the changed factor;
identify the outcome measurement or derived outcome;
identify every control-check measurement;
identify any baseline;
state what each control check protects against;
identify whether any control failed;
write the evidence-supported conclusion;
explain the mechanism separately.

Answer-Checking Receipt

Did I start from the scientific question?
Did I identify the deliberately changed factor?
Did I identify which measurement actually answers the question?
Did I distinguish baseline and control-check measurements?
Did I keep units and object labels attached?
Did I notice any failed or drifting control?
Did I avoid treating every number as an outcome?
Did I use control-check evidence to judge fairness rather than to replace the outcome?
Did I separate evidence from causal explanation?

Parent and Tutor Teaching Guide

When a child sees a method full of numbers, ask them not to calculate anything yet. Ask:

“Which number answers the scientific question, and which numbers are checking that the test stayed fair?”

Then reverse the question. If temperature was previously a control check, make temperature the tested factor in a new investigation. The child should reassign the role instead of memorising “temperature = controlled variable”.

Next introduce a failed control check. Ask whether it changes the outcome value, the fairness of the interpretation, or both. This develops method evaluation rather than vocabulary matching.

Fade the labels over time. The learner is independent when they can read a new method and determine each measurement’s job from the scientific question.

Useful Internal Routes

Authoritative and Teaching-Evidence References

The measurement-role labels in this guide are learner scaffolds for scientific inquiry. They do not replace the wording of a specific school task or create a universal PSLE marking template. The investigation question and evidence decide the role.

The Quiet Return

A number can answer the question.

Another number can make the answer trustworthy.

They are both measurements. They are not doing the same job.

Start with the scientific question. Then every measurement can take its proper place: what we changed, what we measured as the response, and what we checked so the comparison remained worth believing.

Small Group Tutorials