PSLE Science Reality Lab: What Evidence Does an “X Causes Y” Headline Need?

Series ID: PSLE-SCI-REALITY-0004

Wait, what? Two things can change together even when neither one caused the other.

A headline often compresses a complicated evidence trail into a simple sentence: “X causes Y.” That sentence is easy to remember because it has a clear direction. One thing acts. Another thing changes. But scientific reasoning has to earn that arrow.

A Primary 5 or Primary 6 learner already knows an important part of this problem. In an investigation, changing one condition and observing an outcome is not enough unless the method lets you connect the change to the outcome. Real-world headlines can be harder because the method is often hidden and several conditions are changing at once.

Quick Answer

When a headline says “X causes Y”, check five things before accepting the causal arrow.

Pattern: What relationship was actually observed?
Time order: Did X happen before Y?
Comparison: Is there a suitable condition in which X differs while other important conditions are comparable?
Alternatives: Could another factor cause both X and Y, or could Y influence X?
Test: Is there evidence from a method that deliberately changes or controls the suspected cause?

A pattern can be real without proving the cause. Science becomes stronger when the method rules out competing explanations and the conclusion stays within the evidence.

The Exact Learner Job This Reality Lab Owns

This article owns one job: using PSLE Science inquiry to evaluate a real-world causal headline by separating an observed association from evidence that one factor actually produces a change in another.

It is not a statistics chapter and does not replace existing pages on fair tests, changed conditions, mechanisms or alternative explanations. It applies those micro-skills to a communication object that often hides the experimental structure: the causal headline.

Useful routes include telling a data pattern from a scientific mechanism, generating more than one scientific possibility, improving an investigation without changing its question, and keeping conclusions within tested limits.

Reality Lab Case: The Open Window Headline

Imagine a fictional article with this headline:

Opening Windows Causes Laundry to Dry Faster.

The article says researchers observed twenty homes. In homes where a window was open, damp cloths dried in an average of 80 minutes. In homes where windows were closed, the cloths dried in an average of 130 minutes.

The pattern is clear: the open-window homes in this observation had shorter drying times. Is the headline therefore proved?

Not yet. The pattern is evidence. The cause requires more work.

Step 1: State the Pattern Without Adding a Cause

A careful first statement is:

In the observed homes, cloths dried faster on average when a window was open than when windows were closed.

That sentence describes what was observed. It does not yet tell us why.

This is a powerful scientific habit: describe the relationship first; explain the relationship second.

Step 2: Ask Whether the Suspected Cause Came First

A cause must occur before the effect it produces. If a window was opened only after the cloth was already mostly dry, the time order would not support the claim that opening the window caused the earlier drying.

In many headlines, time order is not stated clearly. The data might have been collected at one moment, or the supposed cause and effect might have been measured together.

Time order does not prove causation, but causation without correct time order makes no sense.

Step 3: Keep Alternative Explanations Alive

Why might open-window homes have faster drying even if the window itself is not the only cause?

Open-window homes may also have stronger natural airflow.
They may receive more direct sunlight.
Residents may open windows on less humid days.
The cloths may be hung nearer the opening.
The rooms may be larger or warmer.
The starting amount of water in the cloths may differ.

Any factor that changes together with the suspected cause and can affect the outcome is a competing explanation.

The point is not to create endless doubt. It is to ask whether the evidence distinguishes the preferred explanation from plausible alternatives.

Step 4: Could the Direction Be Reversed?

Sometimes a relationship can be read in two directions.

Imagine a headline says:

Large Leaves Cause Plants to Receive More Water.

Perhaps the data only show that plants with larger leaves were also the plants receiving more water. It is possible that more water contributed to larger leaves rather than larger leaves causing people to give more water.

When two variables move together, ask whether the proposed direction is supported by the sequence and mechanism.

Step 5: What Experiment Would Strengthen the Causal Claim?

Return to the window and drying case.

A stronger investigation could use the same room or closely matched rooms, similar cloths with the same starting water content, the same hanging position, and comparable temperature and humidity. The window condition could be deliberately changed while other relevant conditions are kept as comparable as practical.

If repeated trials show shorter drying times with the window open under these controlled conditions, the causal claim becomes stronger.

Notice the shift: we moved from observing a naturally occurring difference to deliberately testing the suspected cause.

A Mechanism Helps, but a Plausible Mechanism Is Not the Same as a Test

You may know that moving air can help water evaporate from a wet surface. That makes the window explanation scientifically plausible.

But a plausible mechanism cannot repair a poor comparison. If the open-window homes also had much warmer rooms, we still need evidence separating the effects.

Science uses mechanisms and observations together. Mechanism tells us whether a cause makes sense; experimental design helps tell us whether that cause explains the observed difference in this case.

The Third-Factor Pattern

Sometimes X and Y move together because a third factor affects both.

Imagine a fictional school garden survey finds that plots receiving more sunlight also contain taller plants. A headline says:

Sunlight Causes Taller Plants.

The headline may point toward a real biological mechanism, but in the survey the sunnier plots may also receive more water because they are closer to a tap, have different soil, or contain a different plant variety.

A third factor can create or enlarge a pattern. A causal test should make those alternatives less plausible.

A Pattern Can Be Useful Before the Cause Is Known

Do not make the opposite mistake and treat association as worthless.

A repeated relationship can be scientifically useful even before causation is established. It can help researchers generate hypotheses, identify conditions worth testing, make predictions, or find places where a mechanism might operate.

The discipline is to label the evidence correctly. “These two factors were associated in the observations” is different from “changing X will cause Y to change.”

Headline Compression: Where the Extra Certainty Enters

Suppose the full study result is:

In the observed homes, open-window conditions were associated with shorter drying times.

A headline writer may shorten that to:

Open Windows Make Laundry Dry Faster.

The second sentence is shorter and more vivid. It also adds causal certainty unless the method supports it.

Reality Lab asks you to compare the communication verb with the evidence:

was linked with describes a relationship;
was associated with describes a relationship;
occurred together with describes a relationship;
caused, made or led to claims a direction and mechanism.

Worked Transfer Case: Cooler Streets and More Trees

A fictional infographic compares ten streets. Streets with more trees have lower afternoon temperatures. The caption says, “Trees make streets cooler.”

The pattern is compatible with mechanisms such as shading and evaporative cooling. But before assigning the whole difference to trees, check other differences:

street width;
building height;
amount of dark road surface;
traffic;
distance from large open areas;
measurement time and sensor position.

A stronger causal claim could combine observational patterns with controlled or carefully matched measurements that isolate tree cover more effectively.

The scientific response is not “trees do not cool streets”. It is “the infographic alone may not tell us how much of the difference is caused by tree cover rather than other street differences.”

Worked Transfer Case: Clearer Water and More Filter Layers

Five student-made filters are displayed at a fair. Filters with more layers produce clearer-looking water. A poster says, “More layers cause cleaner water.”

Possible problem: filters with more layers also contain more total material and use different material combinations. “Number of layers” is not the only changed factor.

A better test would define the outcome, such as a suitable measure of clarity for the learning task, then vary layer number while keeping material type, total dimensions, starting water and procedure comparable where the question requires it.

The existing poster shows a pattern. A causal investigation must decide which feature of the filters is actually being tested.

Worked Transfer Case: More Practice, Higher Scores

A fictional school graph shows that students who completed more Science practice questions also tended to score higher. Someone writes, “Doing more questions causes higher Science scores.”

The graph alone does not establish the whole causal claim. Students who practise more may also differ in prior knowledge, support, study time or motivation. High-performing students might choose to practise more. The number of questions may also matter less than the quality of practice.

The pattern can still motivate a useful hypothesis. It simply needs a stronger design before the arrow is treated as settled.

What Evidence Would Strengthen an “X Causes Y” Headline?

X clearly occurs before the change in Y.
The method deliberately changes X or uses a strong comparison where X differs.
Other important conditions are kept comparable or measured and considered.
The outcome Y is measured consistently.
The pattern repeats.
A plausible mechanism connects X to Y.
Alternative explanations become less able to account for the result.
The conclusion stays within the tested system and conditions.

What Would Weaken It?

X and Y are measured at the same time with no time order.
Several important conditions differ between groups.
The direction could plausibly run from Y to X.
A third factor could affect both.
The measured outcome is only a rough proxy for the claimed effect.
The headline is broader than the studied objects, times or conditions.

The Causal-Headline Checklist

Rewrite the headline as a testable question.
State the observed relationship without using a causal verb.
Check time order.
Name at least two alternative explanations.
Ask what comparison would isolate the proposed cause.
Identify what outcome should be measured.
Ask whether a mechanism supports the direction.
Write the narrowest conclusion the evidence can carry.

Practice 1: “Brighter Light Makes Leaves Bigger”

Ten plants near a window have larger leaves than ten plants farther inside a room. A poster says brighter light caused larger leaves.

Explained answer: The observation shows a relationship between location/light and leaf size. But plant type, watering, temperature and starting size may also differ. A stronger causal test would use comparable plants and deliberately vary light while controlling relevant conditions and measuring leaf size in the same way.

Practice 2: “More Airflow Causes Faster Evaporation”

Two wet cloths are placed in different rooms. The cloth in the breezier room dries first.

What is missing? Temperature, humidity, starting water content, cloth size, material and other conditions may differ. The result is consistent with airflow affecting evaporation but does not isolate airflow in this two-room comparison.

Practice 3: “Larger Pots Cause Taller Plants”

A nursery record shows that taller plants are usually in larger pots.

Alternative direction: Taller, older plants may have been moved into larger pots because they were already larger. The pot size may be a consequence of plant size rather than the original cause of it.

Delayed Independent Return

In two days, invent a headline of the form “X causes Y”. Then create two evidence packets:

Packet A: only an observed relationship between X and Y;
Packet B: a fairer investigation that deliberately tests X while controlling relevant alternatives.

Write one sentence for what Packet A supports and a stronger—but still bounded—sentence for Packet B. If the two sentences are different, you understand why the causal arrow must be earned.

Common Misconceptions

“If two things change together, one must cause the other.”

No. They may share a cause, move together by coincidence, or influence one another in a different direction.

“If there is a good mechanism, causation is proved.”

No. A mechanism makes the claim plausible, but the particular evidence still needs to isolate the cause sufficiently.

“Association means the result is useless.”

No. Associations can reveal patterns, support predictions and generate hypotheses. The mistake is labelling them as stronger causal evidence than they are.

“An experiment proves the cause forever.”

An experiment supports a causal conclusion within its design and conditions. Different systems or conditions may require new evidence.

Evidence Boundaries and Model Limits

Real-world causal questions can be difficult because it may be impossible or inappropriate to control every factor. Scientists often combine different kinds of evidence: observations, controlled experiments where possible, repeated measurements, mechanisms and studies under varied conditions.

Primary learners do not need to master every research design. They need the durable habit: the stronger the causal verb, the clearer the evidence route must be.

How This Connects Back to PSLE Science

The current PSLE Science assessment objectives include interpreting and analysing information, evaluating observations, information and methods, and communicating explanations and reasoning. Those are exactly the skills required to decide whether a pattern in a table or graph supports a descriptive conclusion, a prediction or a causal explanation.

The world often removes the question stem and gives you the conclusion first. Reality Lab restores the missing scientific job.

Parent and Tutor Teaching Guide

Avoid starting with the phrase “correlation is not causation” as a slogan. A child can memorise it without understanding it.

Instead, use a three-stage progression:

Show a simple relationship and ask the learner to describe it without a causal verb.
Ask for two other explanations that could produce the same pattern.
Ask how to change the method so the proposed cause is tested more directly.

For a stronger learner, reverse the exercise. Give a well-controlled investigation and ask them to write one accurate headline and one overconfident headline. Then identify the exact word where the overstatement enters.

Where to Go Next

Authoritative Sources

Singapore Examinations and Assessment Board, PSLE Science syllabus for examination from 2026.
Singapore Ministry of Education, Science Teaching & Learning Syllabus: Primary, 2023.

Final Return

A causal headline gives you an arrow. Science asks whether the evidence built that arrow or whether the sentence added it.

What was observed, what else could explain it, and what test would make the causal direction harder to deny?

Once you ask those questions naturally, a headline becomes a scientific claim you can examine rather than a sentence you must simply believe or reject.

Small Group Tutorials

PSLE Science Reality Lab Vol No.004 | “X Causes Y” — What Evidence Would a Causal Headline Need?