How to Tell a Scientific Example From Evidence in a PSLE Science Answer

Wait, What? A True Example Can Still Be the Wrong Evidence

A learner is asked how the given results support a scientific conclusion. Instead of using the results, the learner writes a familiar example from notes: “For example, a metal spoon becomes hot in soup.” The example may be scientifically true. It may even illustrate the same concept. But if the question asks what the given evidence shows about the actual set-up, the example has not done the required job.

This distinction matters because PSLE Science answers often combine two different sources of knowledge:

evidence from the question—observations, measurements, comparisons, diagrams, tables or graphs;
scientific knowledge—concepts, relationships and mechanisms you already know.

An example can help explain a concept. It does not automatically become evidence about the case being tested.

AN EXAMPLE SHOWS WHAT A CONCEPT CAN LOOK LIKE. EVIDENCE SHOWS WHAT HAPPENED IN THIS CASE.

Quick Answer

When you are unsure whether a statement in your PSLE Science answer is evidence or an example, use this route:

ASK WHERE THE STATEMENT CAME FROM → ATTACH IT TO THE ACTUAL SCIENTIFIC OBJECT → DECIDE ITS ROLE → IF IT CAME FROM THE QUESTION OR INVESTIGATION, USE IT AS EVIDENCE → IF YOU CREATED IT TO ILLUSTRATE AN IDEA, KEEP IT AS AN EXAMPLE → SELECT THE RELEVANT CONCEPT → EXPLAIN THE MECHANISM → RETURN TO THE ACTUAL EVIDENCE BEFORE STATING THE OUTCOME.

A useful example can support understanding. It cannot replace the observation or measurement the question asks you to use.

The Exact PSLE Science Learning Job This Guide Owns

This guide owns one learner job: distinguishing a scientific example used to illustrate a concept from evidence that actually supports a claim about the object, set-up or investigation in the PSLE Science question.

It does not replace the existing guide on deciding when an answer comes from given evidence versus scientific knowledge, the guide on answering “How do you know?” with evidence, the guide on learning concepts through examples and non-examples, or the guide on using everyday experience without overriding evidence. Those remain separate canonical jobs. This page owns the boundary between illustration and support for the actual case.

Why This Matters in the Current PSLE Science Frame

For examination from 2026, PSLE Science assesses the 2023 Primary Science syllabus. The official assessment objectives include applying scientific facts, concepts and principles; interpreting and analysing information; evaluating observations, information and methods; and communicating explanations and reasoning.

That means a learner must be able to do two things at once: use the information supplied in a question accurately and bring in relevant scientific knowledge when the task requires it. Confusing an example with evidence breaks that connection. The answer may contain correct Science but still fail to respond to the actual information in the question.

Four Roles a Science Statement Can Play

Role	What it does	Typical source
Evidence	Supports a claim about the actual case	Question, observation, measurement, table, graph, diagram, investigation
Concept	Provides relevant scientific knowledge	What you have learned
Example	Illustrates what a concept can look like	You, a textbook, teacher, notes
Explanation	Connects evidence and conditions through a mechanism to an outcome	Your reasoning using evidence + concept

A single sentence can sometimes contain more than one role, but you should be able to identify which part is doing which job.

The Source Test

Ask: where did this statement come from?

If it came from the question’s results, diagram, table, graph, caption, observation or measurement, it may be evidence.
If it came from a scientific principle you learned, it is scientific knowledge.
If you invented a parallel case to make the concept easier to see, it is an example.
If it connects the actual evidence to the actual outcome using a mechanism, it is part of an explanation.

This simple source check prevents a familiar example from quietly becoming a “result” that was never observed.

Worked Example 1 — Heat Transfer: Familiar Example Versus Actual Evidence

Imagine a question gives two rods of different materials. The same amount of one end of each rod is heated. After a stated time, the temperature at the other end of Rod P is higher than at the other end of Rod Q.

The question asks: “How do the results support the conclusion that material P transfers thermal energy more readily under these conditions?”

A weak answer says: “For example, metal spoons get hot in soup.” That may illustrate conduction, but it does not use the actual result.

A stronger evidence answer begins with the comparison supplied: the far end of P reached a higher temperature than the far end of Q after the same stated heating period under comparable conditions. Then scientific knowledge can connect that result to the conclusion.

Worked Example 2 — Plant Response: Do Not Replace the Given Conditions With a Textbook Example

A question describes two plants under specifically stated light conditions and gives measured results. The learner remembers a textbook picture of a plant bending toward a window and writes about that instead.

The window example may be useful during revision. In the answer, however, the learner must first preserve the conditions and evidence in the actual question. A scientific mechanism should explain those results, not a different remembered scenario.

Worked Example 3 — Classification: A Familiar Animal Does Not Prove the Unknown One Belongs to the Same Group

An unfamiliar organism is described by several observable features. A learner says, “A butterfly has wings and is an insect, so this organism is an insect.”

The butterfly is an example of an insect. It is not evidence about the unknown organism. The evidence is the unknown organism’s own observable characteristics. Classification must follow those characteristics and the stated classification criteria.

Worked Example 4 — MCQ Distractor: A True Example Can Make a Wrong Option Feel Right

An answer option contains a general statement that reminds the learner of a familiar example. The learner accepts it without checking whether the question’s exact condition matches the example.

This is why examples can be powerful distractors. They create familiarity. The correct route is to return to the stem: what object is being discussed, under what condition, and what evidence is supplied? Only then test whether the statement applies.

Worked Example 5 — Model Answer Example Versus Your Question

A model answer explains a scientific relationship using one worked example. A learner memorises the example and inserts it into every similar question. Eventually the new question changes an important condition, but the learner keeps the old example anyway.

A model answer is a learning resource, not fresh evidence for another question. Extract the concept and reasoning structure; do not carry the old objects, data or conditions into a new case unless the new question actually supplies them.

Worked Example 6 — When an Example Becomes Part of the Evidence

Sometimes a question itself gives an example as data. For instance, it may state that Object A has a particular characteristic and belongs to Group X, then ask you to use that information to classify another object. In that case, the example is not merely something you invented—it is question-given information and may legitimately function as evidence or a reference case.

The source matters. The same sentence could be “just an example” in your revision notes and “given evidence” when the examination question explicitly supplies it.

Example, Analogy and Evidence Are Not the Same

Three learning tools can look similar:

Example: another case where the concept applies.
Analogy: a comparison that makes one relationship easier to understand.
Evidence: information that supports a claim about the actual case being investigated.

An analogy can make a mechanism memorable without proving that the real system behaves identically in every detail. An example can show application without proving what happened in the question. Evidence is tied to the actual claim.

The Evidence–Concept–Mechanism Chain

READ GIVEN INFORMATION → IDENTIFY THE SCIENTIFIC OBJECT → STATE THE RELEVANT OBSERVATION / MEASUREMENT → DISTINGUISH IT FROM ANY EXAMPLE YOU REMEMBER → SELECT THE RELEVANT CONCEPT → EXPLAIN THE CAUSAL MECHANISM → CONNECT TO THE QUESTION CONDITION → STATE THE OUTCOME → CHECK THAT THE EVIDENCE STILL BELONGS TO THIS CASE.

A Three-Question Role Check

Source: Where did this statement come from?
Role: Is it evidence, concept knowledge, an example or explanation?
Relevance: Does it refer to the actual object and condition in this question?

If you cannot answer these three, the sentence may be doing the wrong job in your response.

Evidence Answers “What Happened Here?”

Evidence can include:

a measured value;
a difference between matched set-ups;
a pattern across a table;
a labelled feature in a diagram;
a repeated observation;
a result after a stated condition changes;
a directly supplied comparison;
a question-given reference case.

Its strength depends on how well the method, comparison and conditions support the claim.

Examples Answer “What Else Could This Idea Look Like?”

Examples are excellent for learning. They can:

make an abstract concept concrete;
show where a definition applies;
contrast with a non-example;
help retrieve a mechanism;
support transfer to unfamiliar contexts;
reveal which features are essential and which are decorative.

But when the task asks you to justify a claim from the supplied evidence, the example should not displace the data.

When an Example Is Useful Inside an Answer

Some learning or communication tasks genuinely ask for an example, or allow one to clarify a scientific idea. If the question says “give an example,” then the example is the requested response job. If a parent or tutor asks you to explain a concept in your own words, a carefully chosen example can demonstrate understanding.

The important rule is not “never use examples in Science answers.” It is: use an example only when it serves the requested job, and do not mistake it for evidence that the question never supplied.

Observable Failure Signatures

Failure signature	Likely weak link
The answer begins “For example…” when the question asks what the data show	Example substituted for evidence
A remembered textbook case replaces the given object	Actual-case identity lost
A familiar example is treated as proof of a causal relationship	Illustration promoted to evidence
The learner invents an observation that would make the answer work	Example turned into fake data
A “How do you know?” question receives only a mechanism	Evidence job missed
The learner quotes all table values and then adds an unrelated example	Evidence not selected; example padding added
The example matches the topic but not the question condition	Condition tracking failed

Find the Earliest Weak Link

What exactly is the question asking: evidence, explanation, example, comparison, conclusion or prediction?
What scientific object does the question own?
What observations or measurements belong to that object?
Which statement in my answer came from the question?
Which statement came from my memory?
Did I introduce an example that was never observed?
Does the example match every important condition?
Could I remove the example and still justify the actual claim from the evidence?

Misconception Repair — “If the Example Is True, It Strengthens My Answer”

Not automatically. Relevance matters. A true example can be unnecessary, off-condition or unrelated to the evidence being evaluated. A short answer using the decisive evidence can be scientifically stronger than a long answer filled with true but irrelevant examples.

Misconception Repair — “Evidence Must Be a Number”

No. Evidence can be qualitative: a colour change, a visible structure, a sequence, an observed event, a category, or another descriptive result. The question is whether the information is an observation or record relevant to the claim.

Misconception Repair — “An Example From My Notes Is Scientific Knowledge, So I Can Use It as Evidence”

The example is part of your background knowledge, but that does not make it evidence about the new case. Use the scientific principle behind the example when relevant. Use the new question’s evidence to support the new question’s claim.

Misconception Repair — “Examples Are Bad in Science”

No. Examples are powerful learning tools. They become a problem only when they are assigned the wrong role. Learn with examples. Reason from evidence. Use each where it belongs.

Question-Reading Protocol: Source → Role → Relevance

Source: label each important statement as QUESTION / SCIENCE KNOWLEDGE / MY EXAMPLE.
Role: decide whether the current task needs evidence, a mechanism, a comparison, an example or something else.
Relevance: attach every statement to the actual object and condition.
Evidence first: when the task asks for support, select the decisive observation or comparison.
Concept next: bring in scientific knowledge only as needed.
Mechanism: explain how the condition leads to the outcome.
Check: remove any decorative example that does not do a required job.

Practice Sequence 1 — Sort the Sentences

Create twelve short Science statements and sort them into four groups: evidence, concept, example and explanation. Include difficult cases where an example is supplied by the question and therefore becomes part of the information available.

Do not memorise surface phrases such as “for example”. The role depends on the source and the task.

Practice Sequence 2 — Replace the Example With Evidence

Take an answer that says:

“This happens because of concept X. For example, familiar case Y does the same thing.”

Now rewrite it around the actual evidence:

“In this question, the evidence is ___. Under condition ___, concept X means ___. Therefore the observed outcome is ___.”

This is a practice frame, not a compulsory examination sentence structure.

Practice Sequence 3 — Turn an Example Into a Question

Choose a familiar scientific example from your notes. Then ask: what evidence would I need if I wanted to prove that the same relationship occurs in a new case?

This reverses the usual direction. Instead of treating the example as proof, you design the observation or comparison that would actually support the claim.

Unfamiliar Transfer Challenge

Take one concept from each of two different Primary Science themes. For each concept:

write one familiar example;
write one new imaginary investigation;
create one observation or measurement from that investigation;
state which sentence is the example and which is the evidence;
build an explanation that uses the evidence and concept without needing the familiar example.

If you can swap the example but preserve the reasoning from evidence to mechanism, the boundary is becoming clear.

Delayed Independent Return Test

Three to five days later, take four unfamiliar Science answers—two containing real evidence and two containing only illustrative examples. Without notes, identify which ones actually support the claim about the case. Then repair the weaker answers by inserting the relevant evidence.

Answer-Checking Receipt

I know what the question is asking me to produce.
I can identify the actual scientific object or set-up.
I know which statements came from the question.
I know which statements came from my prior knowledge.
I can tell an illustrative example from evidence about the case.
I do not invent observations to make an answer work.
I use scientific concepts to interpret evidence, not to replace it.
My explanation returns to the actual condition and outcome in the question.

Common Traps

using “for example” whenever you cannot find the evidence;
copying a familiar textbook scenario into a different condition;
treating everyday experience as proof;
writing an example instead of a relationship statement from the data;
adding examples as padding to an already complete answer;
confusing a reference case supplied by the question with an example you invented;
turning a model answer’s objects and numbers into permanent facts.

Parent and Tutor Teaching Guide

When a child gives a familiar example, do not immediately say it is wrong. Ask: “Is that what happened in this question, or is that another case that helps you remember the concept?” This preserves the value of the example while making the role distinction visible.

Then point to the table, diagram or observation and ask, “What evidence belongs to this case?” Once the learner identifies it, ask which scientific idea helps explain it. This naturally builds the evidence → concept → mechanism chain.

Use paired practice. In one question explicitly ask for an example; in the next ask for evidence supporting a conclusion. Keep the scientific topic similar. The learner must change the response job even though the topic stays familiar.

When reviewing model answers, cover the wording and ask the learner to label each sentence: EVIDENCE / SCIENTIFIC KNOWLEDGE / MECHANISM / OUTCOME. If an example appears, ask whether it was requested and what job it serves.

Useful Internal Routes

Authoritative References

Singapore Examinations and Assessment Board — PSLE Science syllabus, for examination from 2026
Ministry of Education, Singapore — Science Teaching & Learning Syllabus, Primary, 2023
National Research Council — A Framework for K–12 Science Education, used as broader science-practice context rather than PSLE examination policy.

Evidence and Boundary Note

This guide does not claim that “example versus evidence” is a named PSLE marking category or that examples are forbidden in examination answers. The learning distinction is about source and role. When a task asks you to use evidence, use information that belongs to the actual case. When a task asks for an example, give an appropriate example. When an explanation is required, connect the relevant evidence and scientific knowledge to the stated outcome.

The Quiet Return

Examples help you see the Science.

Evidence helps you decide what happened here.

World-class reasoning knows the difference—and knows when each one belongs in the answer.

Small Group Tutorials