Primary 3 Science Learning Guide | Same Test, Different Question: What Does the Evidence Answer?

The same experiment can support several different Science questions—but only if the pupil understands what the evidence actually measures.

Primary 3 pupils often become comfortable with a test and then assume every question about that setup has the same answer. But one set of results can be used to describe what happened, compare samples, infer a property, make a prediction, explain suitability or identify what remains unknown.

This guide develops a powerful scientific habit: separate the test from the question being asked about the test. The evidence stays the same; the reasoning job can change.

Wait, What? Same Evidence Does Not Mean Same Answer

A material strip bends farther than another before breaking.

Describe: Sample A bent farther before breaking.
Compare: Sample A bent farther than Sample B.
Infer: Sample A is more flexible under the test.
Justify: Sample A may be more suitable for a bendable strap because it is more flexible.

The evidence is unchanged. The command determines the answer structure.

Start With: What Was Actually Tested?

Before answering, identify the property or relationship the method was designed to investigate.

Bending test → flexibility.
Load test → strength.
Water absorption test → waterproofness.
Visibility-through-material test → transparency.
Float/sink test → floating behaviour.
Magnet attraction test → magnetic response.
Repulsion with a known pole → strong evidence of magnet behaviour.

A correct answer cannot outrun the test. If the method did not investigate strength, strength should not suddenly appear in the conclusion.

Question Type 1: Describe

Describe asks what happened or what is shown. Stay close to the observation.

Example: “Sample B did not absorb water.”

Do not automatically turn the answer into an explanation unless the command asks for one.

Question Type 2: Compare

Compare asks for similarity or difference using the same basis.

Example: “Sample A bent farther than Sample B before breaking.”

If both floated, a similarity can be stated. If one floated and one sank, the difference can be stated.

Question Type 3: Infer

Infer moves from the observation to the scientific meaning.

Observation: Sample A bent farther before breaking.

Inference: Sample A is more flexible.

The inference should be the property or relationship that the test actually supports.

Question Type 4: Predict

Predict uses a pattern or rule to state what is expected under matching conditions.

If like poles repelled in known examples and a new setup has N facing N, the pupil can predict repulsion.

If a material repeatedly did not absorb water under the same simple test, the pupil may predict a similar result when the test is repeated in the same way. The prediction should not become “it can never absorb water under any possible condition”.

Question Type 5: Explain

Explain connects the evidence to the scientific relationship.

Example: “The magnets repel because the facing poles are both North, and like poles repel.”

A good explanation includes the condition and the rule, not just the final result.

Question Type 6: Justify a Choice

Justify asks why one option is more suitable. This often adds a function to the evidence.

Evidence: Material C is waterproof and flexible.

Function: the cover must bend and keep water out.

Justification: “Material C is suitable because it is flexible and waterproof, so it can bend without breaking while helping keep water out.”

Question Type 7: What Can You Not Conclude?

This question checks evidence limits.

If a sample floated, we can conclude that it floated in the test. We cannot conclude from that result alone that it is waterproof, flexible, strong or transparent.

Knowing what the evidence does not answer is part of understanding the test.

One Test, Several Valid Questions

Suppose three material samples receive the same amount of water.

A absorbs water.
B does not absorb water.
C does not absorb water.

From this one test:

Describe: B and C did not absorb water.
Compare: A absorbed water while B and C did not.
Infer: B and C are waterproof under the test.
Choose: B or C may be suitable if waterproofness is the only required property.
Limit: the test does not tell which is more flexible.

Worked Example 1: Strength Test

Sample A supports a heavier load than B before breaking.

Describe: A supported the heavier load.
Compare: A supported more load than B.
Infer: A is stronger under the test.
Justify: A may be more suitable for a support if strength is the required property.
Limit: the test does not show which is more flexible.

Worked Example 2: Magnet Attraction

Unknown Bar X is attracted to a known magnet.

Describe: X moved toward the magnet.
Infer: X shows magnetic attraction.
Can we conclude X is definitely a magnet? No.
Why not? A magnetic material can also be attracted.
Next useful question: Can X repel a known pole?

The same observation supports a limited conclusion and reveals a need for a better test.

Worked Example 3: Magnet Repulsion

A known N pole repels the left end of X.

Describe: the magnets moved apart.
Infer: the facing end of X is also N.
Explain: like poles repel.
Further infer: the opposite end of X is S.
Conclusion: X behaves as a magnet.

One test can therefore produce a chain of increasingly specific conclusions when the evidence supports them.

Worked Example 4: Life-Cycle Diagram

Diagram: egg → larva → pupa → adult.

Identify: the stage after larva is pupa.
Describe: the cycle has four major stages.
Compare: it differs from a nymph cycle because it includes larva and pupa.
Predict: if the pupa is shown, adult comes next.
Limit: the diagram does not show how long each stage lasts unless time data is added.

Worked Example 5: Living-Thing Evidence

An unknown organism grows and reproduces.

Describe: it became larger over time and produced young.
Infer: it shows growth and reproduction.
Classify: the evidence supports that it is living.
Explain: growth and reproduction are characteristics of living things.
Limit: the evidence may still be insufficient to determine whether it is plant, animal, fungus or bacterium.

Worked Example 6: Float/Sink Test

Sample P floats while Q sinks.

Describe: P stayed at the surface; Q sank.
Compare: P floated while Q sank.
Choose: P is supported if the task requires a sample that floats.
Limit: no conclusion about transparency or waterproofness from this test alone.

The Command Changes the Answer, Not the Evidence

This sentence is worth remembering. Pupils often change the scientific facts when the command changes. They should not.

The evidence remains stable. The answer changes in how it uses the evidence.

One Evidence Set Can Have Different Levels of Conclusion

Some conclusions are direct observations. Others are inferences. Others add a practical decision.

Observation → Property/Pattern → Explanation → Application.

Pupils should know which level the command requires.

Do Not Turn “Infer” Into “Invent”

An inference must be supported by the evidence. If a material is waterproof, the pupil cannot infer that it is also strong unless a strength test was performed or strength information was given.

Inference is evidence-based interpretation, not permission to guess.

Do Not Turn “Explain” Into “Repeat”

If the question says, “Explain why the magnets repel,” writing “They repel because they move apart” repeats the observation.

A real explanation gives the relationship: “They repel because like poles are facing.”

Do Not Turn “Justify” Into “Name a Property”

“Material A is waterproof” may be incomplete if the question asks why it is suitable.

Complete the property-function link: “Material A is waterproof, so it does not absorb water and helps keep the contents dry.”

The Same-Test Reasoning Routine

Test: What property or relationship did the method investigate?
Evidence: What actually happened?
Command: Describe, compare, infer, predict, explain or justify?
Answer level: Observation, concept, relationship or application?
Limit: What remains unknown from this test?

Common Mistakes

Using the same sentence for every command word.
Inferring properties the test did not measure.
Repeating observations when explanation is required.
Giving an explanation when only description is required.
Ignoring evidence limits.
Assuming one successful test answers every property question.
Forgetting the function when justifying suitability.
Changing the Science because the wording of the command changed.

How to Practise

Use one simple setup and ask five questions from it. For example, give a material-bending result and ask pupils to describe, compare, infer, justify and state one thing the test cannot tell them.

Then repeat the exercise with a magnet setup, a life-cycle diagram and a living-characteristics scenario. The structure transfers across topics.

A Mini Diagnostic

Use one flexibility result to answer describe, infer and justify questions.
Explain what a waterproofness test cannot tell you.
Use attraction evidence to state a safe conclusion and one unsupported conclusion.
Use repulsion evidence to infer a pole.
Use one life-cycle diagram to identify, compare and predict.
Explain why “infer” is not the same as “guess”.
Explain why the command can change while the evidence stays the same.

Primary 3 Science Checkpoint

I identify what the test was designed to answer.
I can describe without over-explaining.
I can compare using the same basis.
I infer only what the evidence supports.
I predict from a known pattern or rule.
I explain by connecting condition and scientific relationship.
I justify by connecting property and function.
I can state what the test does not tell me.
I understand that one evidence set can support several question types.
I keep the scientific facts stable when the command changes.

Continue the Primary 3 Science Learning Guide

Return to the Primary 3 Science Learning Hub.

Source and Syllabus Alignment

This guide supports the observing, comparing, inferring, predicting, investigating and communicating practices in the Singapore Ministry of Education Science Teaching & Learning Syllabus: Primary Three to Six.

Small Group Tutorials