How to Write a PSLE Science Relationship Statement From Data Without Turning It Into a Cause

Wait, What? “They Change Together” Does Not Automatically Mean “One Caused the Other”

A table shows that as the tested temperature rises, the measured time becomes shorter. A learner writes, “Higher temperature causes the time to decrease.”

That may eventually be the correct scientific explanation. But the data themselves first establish a relationship: across the tested conditions, higher temperature is associated with shorter measured time. The causal mechanism is a second reasoning job.

This distinction matters in PSLE Science because learners are often asked to describe, compare, conclude or explain. A strong answer knows when the evidence supports a pattern and when scientific knowledge is needed to explain why that pattern occurs.

Quick Answer

To write a relationship statement from data, name the two relevant variables or quantities, state how the measured outcome changes as the tested condition changes, keep the claim inside the tested range, and mention exceptions or turning points if they matter. Do not insert a causal mechanism unless the question asks for an explanation and the evidence plus scientific knowledge support it.

The reasoning chain is: READ THE DATA → IDENTIFY THE TESTED CONDITION → IDENTIFY THE MEASURED OUTCOME → ORDER OR ALIGN THE VALUES → STATE THE OBSERVED RELATIONSHIP → CHECK FOR EXCEPTIONS, PLATEAUS OR REVERSALS → LIMIT THE CLAIM TO THE EVIDENCE → THEN, IF ASKED, EXPLAIN THE MECHANISM.

Owned PSLE Science Learning Job

This guide owns one precise job: turning PSLE Science data into an evidence-bounded relationship statement without turning the observed pattern itself into a causal explanation.

It does not replace the existing guides on telling a trend from a single comparison, translating relationships between tables and graphs, or distinguishing a data pattern from a scientific mechanism. Those pages remain canonical for their broader jobs. This page concentrates on the sentence a learner writes after reading the data.

Why This Matters in the Current PSLE Science Frame

The 2026 PSLE Science examination assesses attainment in the 2023 Primary Science syllabus. SEAB includes interpreting and analysing information, evaluating observations and information, and communicating explanations and reasoning. A relationship statement is one way of converting data into a disciplined scientific claim.

The safest sequence is evidence first, explanation second. If the learner jumps directly to a remembered mechanism, the answer may ignore what the actual table or graph shows.

A Relationship Statement Has Four Core Parts

Part	Question	Example language
Tested condition	What was varied or compared?	As the temperature increases…
Measured outcome	What was observed or measured?	…the time taken decreases.
Direction or pattern	How do the values move?	increases, decreases, remains similar, rises then falls
Evidence boundary	Where does this statement apply?	among the tested values / from 20°C to 50°C

A simple relationship sentence is therefore not a slogan. It is a compressed description of the evidence structure.

Worked Example 1: A Simple Monotonic Relationship

Tested condition X	Measured outcome Y
1 unit	18 units
2 units	14 units
3 units	11 units
4 units	8 units

A strong relationship statement is: As X increases from 1 to 4 units, Y decreases from 18 to 8 units.

This says what the data show. It does not claim that X alone caused Y to decrease, and it does not claim the relationship continues forever beyond X = 4.

Worked Example 2: One Pair Is Not Yet a Trend

Suppose you compare only two conditions: X = 1 gives Y = 18, while X = 2 gives Y = 14.

You can state a comparison: Y is lower at X = 2 than at X = 1. Be cautious about claiming a broad decreasing trend unless the question provides enough ordered conditions to support it.

This protects the learner from turning one comparison into a universal relationship.

Worked Example 3: Equal Input Steps Do Not Require Equal Output Steps

X	Y
10	4
20	7
30	9
40	10

As X increases, Y increases. But Y does not rise by the same amount each time. Do not write “Y increases proportionally with X” or “doubling X doubles Y” unless the values actually support that stronger claim.

A relationship can be increasing without being proportional.

Worked Example 4: A Plateau Changes the Sentence

X	Y
1	3
2	6
3	8
4	8
5	8

Do not force one sentence such as “Y increases as X increases” across the whole range. A more accurate statement is: Y increases as X rises from 1 to 3, then remains at 8 units across the tested values from X = 3 to X = 5.

The flat section is part of the evidence. It may later require a scientific explanation, but the relationship statement should first describe it honestly.

Worked Example 5: A Turning Point Needs Two Directions

X	Y
1	5
2	9
3	12
4	10
5	7

Here Y rises and then falls. The correct relationship statement must preserve both sides: Y increases as X rises from 1 to 3, reaches its highest tested value at X = 3, then decreases as X rises from 3 to 5.

A single “increases with X” sentence would erase half the evidence.

Worked Example 6: Categorical Conditions Are Not a Number Scale

Material	Measured change
P	4 units
Q	9 units
R	6 units

P, Q and R are labels, not increasing numerical values. Do not say, “As the material increases from P to R…” There is no scientific numerical order unless the question gives one.

Instead write comparisons: Material Q shows the greatest measured change among P, Q and R.

Relationship Is Not Mechanism

Consider the statement: “As light intensity increases across the tested conditions, the measured output increases.” That describes a pattern.

An explanation would need to identify the relevant scientific process and show why changing the condition leads to the observed outcome. Those are different layers.

In some investigations, a fair experimental design plus scientific knowledge can support causal reasoning. In others, the evidence shows only a relationship. Do not upgrade every pattern into a cause simply because Science often asks “why”.

Observation → Relationship → Explanation

Layer	Example
Observation/data	At X = 1, Y = 18; at X = 4, Y = 8
Relationship statement	As X increases across the tested values, Y decreases
Explanation	A relevant scientific mechanism connects the changed condition to the outcome

A learner who can move through these layers without merging them is much less likely to restate a graph as though it were an explanation.

The Tested-Range Rule

If data were collected from 20°C to 50°C, the safest relationship statement is about those tested values. “As temperature increases from 20°C to 50°C…” is stronger scientifically than “As temperature increases, this always happens.”

Outside the tested range, a threshold, plateau, reversal or different mechanism may appear. Prediction beyond the data needs additional reasoning.

How to Use Numbers Without Copying the Whole Table

Sometimes one or two decisive values make the relationship clearer. You do not need to rewrite every row.

For example: As X increased from 1 to 4 units, Y decreased from 18 to 8 units. This captures the overall relation compactly.

If there is an exception, turning point or plateau, include the values that reveal it. Select evidence for the claim rather than copying data mechanically.

Worked Example 7: A Single Odd Value

X	Y
1	4
2	7
3	6
4	11
5	14

The overall pattern tends upward, but the value at X = 3 is lower than at X = 2. Do not pretend the data rise at every step.

A careful statement is: Y generally increases as X increases across the tested values, although Y decreases from X = 2 to X = 3.

What the unusual point means is a separate question. It may reflect ordinary variation, a method issue or a real feature of the system. First preserve it.

Worked Example 8: Two Variables Change, So Causal Language Becomes Dangerous

Set-up P receives more light and more water than Set-up Q. P also grows more. A learner writes, “More light caused greater growth.”

The result shows a difference between the set-ups. But because more than one condition changed, the evidence does not isolate light as the sole cause. The relationship statement must not erase the water difference.

This is why fair-test logic matters when moving from relationship to cause.

Earliest Weak-Link Diagnosis

Failure signature	Earliest weak link	Repair
You name only the outcome, not what changed with it.	Variable pairing	Name both tested condition and measured outcome.
You claim a broad trend from two data points.	Evidence breadth	State a comparison unless more ordered values support a trend.
You write “causes” when the data only show co-change.	Pattern vs mechanism	Write the relationship first; explain causation separately if justified.
You ignore a plateau or turning point.	Data structure	Split the statement into regions.
You invent a numerical order for categories.	Variable type	Compare categories directly rather than forcing a trend.
You claim the pattern continues beyond tested values.	Scope	Bind the statement to the tested range.

Misconception Repair: A Relationship Sentence Is Not a Weaker Explanation

Students sometimes think an answer sounds “more scientific” if it uses the word because. But adding a mechanism where none is requested can make the answer less accurate.

Scientific strength comes from matching the claim to the job. If the question asks what the data show, describe the relationship. If it asks why, then connect the verified pattern to the relevant concept and mechanism.

A Reliable Sentence Builder

Use this as a temporary scaffold:

As [tested condition] changes from [range], [measured outcome] [increases/decreases/remains similar/rises then falls], as shown by [decisive values if useful].

Do not memorise this as a compulsory examination phrase. Its purpose is to make the scientific roles visible until the learner can write naturally and accurately.

Question-Reading Protocol

What does the question ask: describe, compare, conclude or explain?
Which variable or condition is ordered or changed?
Which quantity is measured?
Is there enough data for a trend, or only a pairwise comparison?
Does the pattern stay in one direction?
Are there plateaus, turning points or exceptions?
Are the conditions numerical or categorical?
What is the tested range?
Does my sentence accidentally claim cause?

Practice Sequence

Round 1: Write simple increasing and decreasing relationship statements from four-row tables.
Round 2: Separate a single comparison from a multi-point trend.
Round 3: Practise plateaus and turning points.
Round 4: Mix numerical and categorical conditions.
Round 5: For each data set, write one relationship sentence and then, separately, one mechanism sentence only when justified.

Unfamiliar Transfer Challenge

A fictional device produces these results: at input levels 2, 4, 6, 8 and 10, the output values are 3, 6, 9, 9 and 8.

A strong description is: Output increases as input rises from 2 to 6, remains unchanged from input 6 to 8, then decreases when input rises to 10.

You do not need to know the device’s mechanism to describe the relationship correctly. That is the point of transfer.

Delayed Independent Return Test

Several days later, give one unfamiliar graph, one table with a plateau, one categorical comparison and one data set with a turning point. Remove all topic headings.

The learner should identify the two quantities, state the relationship, preserve the tested range and avoid causal language unless a separate explanation is requested.

Answer-Checking Receipt

Did I name both scientific quantities or variables?
Did I describe the actual direction in the data?
Am I claiming a trend from enough evidence?
Did I preserve plateaus, reversals or exceptions?
Did I avoid forcing an order onto categories?
Did I keep the claim inside the tested range?
Did I accidentally turn a relationship into a cause?
If an explanation is required, did I add the mechanism as a separate reasoning step?

Common Traps

Writing only “Y increased” without saying what changed alongside it.
Calling any two-point difference a trend.
Using “directly proportional” because both quantities rise.
Ignoring a plateau or turning point.
Inventing an ordered trend across categories.
Writing “X causes Y” when several conditions changed.
Extending the pattern beyond the measured range.

Parent and Tutor Teaching Guide

Ask the learner to read a graph twice. On the first pass, allow only relationship language: increase, decrease, same, highest, lowest, turning point, plateau. On the second pass, ask whether the question also requires a scientific explanation.

If the learner uses “because” too early, ask, “Which part of that sentence came from the data, and which part came from your Science knowledge?” This separates evidence from mechanism without discouraging explanation.

Use fictional data occasionally. When the learner cannot rely on a familiar topic, the structure of relationship reasoning becomes visible.

Useful Internal Routes

Authoritative References and Evidence Boundary

The sentence scaffolds in this guide are not official marking phrases. They are tools for preserving variables, direction and evidence scope. More advanced scientific data analysis can require statistical modelling and uncertainty treatment beyond the Primary Science level.

The Quiet Return

Data do not explain themselves, but they do deserve to be described accurately before explanation begins.

Name what changed. Name what was measured. State the relationship the evidence actually shows. Then, when the question asks why, bring in the scientific mechanism without pretending the pattern was already the cause.

Continue learning: Return to PSLE Science: reasoning, relationships and causation, or choose English, Mathematics and Science.

Compare this evidence boundary with PSLE English: Meaning Under Uncertainty; text-based inference and scientific causation remain different tasks.

Small Group Tutorials