How to Identify What Evidence a PSLE Science Question Actually Gives You

Wait, What? The Question Can Contain the Answer Route Without Containing the Answer

A Primary 5 or 6 learner can know the Science concept perfectly and still answer the wrong question. That happens when memory takes over before the evidence has been read.

A stem may mention a plant, two containers, a graph, a temperature change and a piece of apparatus. Your brain recognises familiar words and rushes towards a familiar explanation. But PSLE Science reasoning is safer when you begin with a different question:

What does this question actually give me as evidence, and what does it merely tempt me to assume?

Quick Answer

Before selecting a concept, separate the information into five evidence types: observations, measurements, comparisons, conditions and relationships. Then ask what each piece can support and what it cannot support. Only after that should you activate the relevant Science concept and build the causal explanation.

The learner job is not “find a keyword”. It is:

READ GIVEN INFORMATION → IDENTIFY THE SCIENTIFIC OBJECT OR RELATIONSHIP → SEPARATE OBSERVATION FROM INFERENCE → SELECT THE RELEVANT CONCEPT → EXPLAIN THE MECHANISM → CONNECT TO THE CONDITION → STATE THE OUTCOME → CHECK AGAINST THE EVIDENCE.

Owned PSLE Science Learning Job

This guide owns one narrow but powerful PSLE Science learner job: extracting the evidence supplied by a question before solving it. It does not replace the canonical concept pages on heat, forces, electricity, plants, materials, cycles or systems. It teaches the learner how to use whatever scientific evidence the question places in front of them.

The Official Frame for 2026

The Singapore Examinations and Assessment Board states that the PSLE Science examination from 2026 assesses attainment in the 2023 Primary Science syllabus. Its assessment objectives include knowledge with understanding, application of scientific facts, concepts and principles, and scientific inquiry involving prediction or hypothesis, interpretation and analysis, evaluation of observations, information and methods, and communication of explanations and reasoning.

That official frame matters here because “evidence extraction” is not a trick. It is part of interpreting and analysing scientific information. The 2023 MOE syllabus also organises Primary Science through the connected themes Diversity, Cycles, Systems, Energy and Interactions, so evidence may require more than one theme to interpret.

The Five Evidence Types

1. Observations

An observation is something the question shows, states or records directly. Examples include “water droplets formed on the outside of the container”, “the bulb was dimmer”, “the spring became longer”, or “the leaf in Set-up B had fewer starch-positive regions after the test”.

Do not immediately convert an observation into a cause. “The bulb was dimmer” is evidence. “The current was smaller” is already an interpretation unless the current was measured.

2. Measurements

Measurements are observations with quantities: 26°C, 12 cm, 4 minutes, 30 mL, 8 leaves, or a graph value. Measurements are powerful because they permit comparisons, but they still do not explain themselves.

A temperature rose from 25°C to 31°C. That is a measured change. The mechanism responsible for the rise must still be reasoned from the system.

3. Comparisons

Many Science questions hide their real evidence in comparison language: higher, lower, faster, slower, more, less, brighter, darker, longer, shorter, greater, smaller, same or different.

A comparison always needs a reference. “Plant A grew more” is incomplete thinking until you ask: more than what, over what time, under which conditions, and according to which measured quantity?

4. Conditions

Conditions tell you what is true in this particular setup. They include temperature, light exposure, whether a switch is open or closed, which surface is rougher, whether the same amount of water is used, whether air can enter, or which organism is present.

A scientific fact can be true in general and still be wrong for the question because the condition is different.

5. Relationships

Sometimes the evidence is not one fact but a pattern: as one quantity increases, another decreases; two variables change together; one stage follows another; a part is connected to another part; one set-up differs from another in exactly one controlled way.

Relationships are often where PSLE Science reasoning begins. The task is to decide which Science concept can explain that relationship without exceeding the evidence.

The Evidence Extraction Protocol

Name the object or system. What are you actually looking at?
Underline only givens. Record what the stem, diagram, table or graph directly supplies.
Circle changed conditions. What differs between set-ups?
Box measured outcomes. What was observed or measured after the change?
Separate observation from inference. Do not write the cause in the evidence column.
Ask what the evidence can support. Is it enough for a comparison, conclusion, explanation or prediction?
Activate the concept. Now choose the relevant scientific principle.
Build the causal chain. Condition → mechanism → outcome.
Return to the evidence. Does every important part of the answer match something the question actually shows?

Worked Example 1 — The Cold Can

A metal can is taken from a refrigerator and placed in a warm room. After several minutes, droplets appear on the outside surface.

Evidence: the can is cold; the surrounding room is warmer; droplets appear outside.

Not evidence: “water leaked through the metal”. That is an inference, and the stem does not establish it.

Relevant concept: water vapour in the surrounding air can condense when cooled sufficiently near the cold surface.

Mechanism: air near the cold can cools; some water vapour condenses; liquid droplets form on the outside.

The important skill is not knowing the word “condensation”. It is noticing that the evidence points to the surrounding air and the cold outer surface, not to an invisible leak.

Worked Example 2 — Two Materials, One Temperature Table

Two identical cups of warm water are wrapped with different materials. The initial temperature is the same. After 15 minutes, Cup A is 58°C and Cup B is 52°C.

Evidence: same starting temperature, same duration, different wrapping material, Cup A ends warmer.

Relationship: Cup A lost less thermal energy to the surroundings over the same period than Cup B, within the limits of the setup.

Possible conclusion: Material A is a better thermal insulator than Material B under these test conditions.

Do not add unsupported claims such as “Material A is always the best insulator” or “Material B conducts heat extremely fast”. The data compare only these two materials in this setup.

Worked Example 3 — A Graph With a Pattern

A graph shows that as the amount of light reaching a plant increases across several tested levels, the measured rate of a process rises and then levels off.

The evidence is the pattern. The graph does not automatically prove that light is the only factor affecting the process. The plateau may indicate that, in that tested range, some other condition becomes limiting, but the graph alone cannot identify which condition unless more evidence is supplied.

This is a crucial scientific habit: read what the graph establishes before explaining what might cause it.

Diagram Evidence: Treat Arrows and Labels as Information, Not Decoration

In PSLE Science, a diagram can contain more evidence than the paragraph. An arrow may show direction, a gap may show a broken connection, a shaded region may indicate light blocked from part of a leaf, and labels may identify the parts that matter.

When a diagram is present, ask:

What do the labels identify?
What does each arrow represent?
What is connected?
What is separated?
What changed between versions of the diagram?
What is schematic and what is meant to be interpreted literally?

Table Evidence: Read Across and Down

A table can hide relationships because the eye focuses on individual numbers. First read the headings and units. Then compare down a column for change across conditions, and across a row for differences at the same condition.

Never quote a number without knowing what quantity it measures.

What the Evidence Does Not Give You

Strong Science also notices missing evidence. A question may not tell you the exact mechanism, may not measure every variable, may show association without proving cause, or may compare only two conditions.

The correct response is not to fill every gap with imagination. Ask whether the missing information is necessary for the requested answer.

Failure Signature 1 — The Familiar-Topic Jump

The learner sees “plant” and immediately writes about photosynthesis. Repair: cover the question prompt and write three evidence statements before naming any concept.

Failure Signature 2 — True Fact, Wrong Evidence

The learner writes a scientifically true sentence but it does not explain the observation in the question. Repair: add the stem-specific bridge, beginning with “In this set-up…” or “Because the question shows…” and then connect to the mechanism.

Failure Signature 3 — Inference Disguised as Observation

The learner writes “the plant photosynthesised faster” as though it were directly observed when the question only reports another measurement. Repair: make two columns: observed and inferred.

Failure Signature 4 — Ignoring the Baseline

The learner says a value is “high” without identifying the comparison. Repair: every comparative word must point to a reference value, earlier state or other set-up.

Failure Signature 5 — Evidence Dumping

Some learners copy every number from a table. That is not analysis. Repair: ask which data point or relationship changes the scientific decision.

Earliest Weak-Link Diagnosis

If a learner repeatedly gets PSLE Science questions wrong, diagnose the earliest failure, not the final sentence.

If the wrong object is identified, repair question reading.
If givens and assumptions are mixed, repair observation/inference separation.
If the wrong comparison is used, repair evidence selection.
If the evidence is correct but the mechanism is wrong, repair the concept.
If the mechanism is correct but disconnected from the condition, repair causal linking.
If the answer is correct but too strong, repair evidence limits.

Retrieval Practice: Evidence First

Take a completed PSLE-style practice question. Hide the answer. Before solving, retrieve only:

three givens;
one changed condition;
one measured outcome;
one relationship;
one thing the question does not tell you.

Then solve. This trains the reading layer separately from the concept layer.

Unfamiliar Transfer Test

Use a made-up device or unfamiliar organism. The learner should still be able to identify the evidence even when they do not recognise the context. If evidence extraction survives a strange surface example, the skill is becoming portable.

Delayed Independent Return Test

Two or three days later, give the learner a new mixed-theme question and no hints. Ask for an “evidence receipt” before the answer:

What is directly observed?
What is measured?
What changed?
What stayed the same?
What relationship appears?
What cannot yet be concluded?

If the learner can do this quickly and accurately, the skill has begun to return independently.

Answer-Checking Receipt

Did I use evidence from the actual question?
Did I separate observation from inference?
Did I choose a concept that explains the evidence?
Did I include the changed condition?
Did I connect cause to outcome?
Did I claim more than the data support?

Useful Internal Routes

Parent and Tutor Teaching Guide

Do not begin correction with “The answer is…”. Begin with “Show me the evidence.” Ask the learner to point to the sentence, diagram feature, table cell or graph pattern that justifies each claim.

If a child answers from memory, remove the topic label. Say: “Pretend you have never seen this example. What does the question physically show?” This prevents memorised chapter associations from outrunning the evidence.

When the learner makes a good inference, praise the connection, not the confidence: “You used the measured difference and connected it to the concept.” That builds a scientific habit rather than a guessing habit.

Authoritative References

The Quiet Return

Evidence is the floor beneath the answer. Concepts tell you what the evidence may mean, but the question decides which concept is relevant and how far the conclusion can go.

Read the world the question gives you first. Then explain it.

Small Group Tutorials