A pupil can know every chapter and still struggle when a Science question shows an unfamiliar experiment.
The reason is simple: an investigation question is not asking only, “Do you remember the topic?” It may also ask, “Can you work out what was changed, what was measured, what was kept comparable, what the results show and how strongly the evidence supports the conclusion?”
Scientific knowledge becomes powerful when the learner can use it to read evidence, not only recite it.
This fourth core guide in the Primary 4 Science Learning Hub develops the investigation-and-transfer layer that connects the whole Primary 4 Science system.
Quick Answer: What Is the Investigation Job?
When reading an investigation, ask:
- What question is being investigated?
- What condition is changed?
- What result is measured or observed?
- What other important conditions should remain comparable?
- How was the measurement made?
- What pattern do the results show?
- What conclusion is supported?
- What conclusion would go beyond the evidence?
A useful eduKate routine is:
QUESTION → CHANGE → KEEP → MEASURE → READ → CONCLUDE → TRANSFER
This is a teaching routine, not an official MOE marking formula.
Why Primary 4 Is a Good Year to Learn This
The current MOE Primary Science syllabus develops scientific practices throughout Primary Science. At Primary 4, learners meet content that naturally supports investigation: measuring mass and volume, investigating variables affecting shadows, measuring temperature, comparing heat conduction and using evidence to validate explanations.
Official reference: MOE Science Teaching & Learning Syllabus — Primary.
This page does not replace a school’s practical-work instructions or marking scheme. Its purpose is to make the reasoning structure visible.
Wait, What? A Fair Test Is Not “Everything Is the Same”
If everything were exactly the same, there would be nothing to investigate.
A fair comparison usually changes the factor of interest while keeping other important conditions sufficiently similar so the outcome can reasonably be related to the changed factor.
That gives three useful roles:
- Changed condition: the factor deliberately altered.
- Measured or observed result: the outcome recorded.
- Controlled conditions: other relevant factors kept comparable.
Different schools may use terms such as independent variable, dependent variable and controlled variable. The simpler language above is useful because it makes the relationship clear before terminology is memorised.
Investigation Example 1: Shadow Distance
A pupil investigates how the distance between an object and a torch affects the width of its shadow on a fixed screen.
The torch and screen stay in the same positions. The same object is moved to three distances from the torch. Shadow width is measured each time.
| Distance from torch | Shadow width |
|---|---|
| 10 cm | 18 cm |
| 20 cm | 13 cm |
| 30 cm | 10 cm |
Changed condition: distance between object and torch.
Measured result: shadow width.
Useful controls: same torch, same object, same screen position, same method of measuring width.
Supported conclusion: Under this arrangement, increasing the object’s distance from the torch was associated with a smaller measured shadow width.
Overclaim: “Any object farther from any light source always makes exactly this size of shadow.” The data do not support that universal statement.
Investigation Example 2: Keeping Water Warm
Two identical cups contain the same amount of water at the same starting temperature. Cup A has no wrapping. Cup B is wrapped in Material X. The cups remain in the same room for 20 minutes.
| Cup | Starting temperature | Temperature after 20 min |
|---|---|---|
| A | 70°C | 50°C |
| B | 70°C | 58°C |
Changed condition: presence of Material X wrapping.
Measured result: final temperature or temperature decrease after the same period.
Evidence: Cup B decreased by 12°C while Cup A decreased by 20°C.
Reasonable conclusion: Under the tested conditions, Material X reduced the water’s temperature drop compared with the unwrapped cup.
Notice that calculating the change gives a clearer comparison than looking only at the final values.
Investigation Example 3: Plant Roots
Two similar seedlings are placed in similar soil under similar light conditions. Both receive the same amount of water. Plant P has an intact root system. Many roots of Plant Q are damaged.
After two days, Plant Q is more wilted.
The question may look like a plant-facts question, but it is also an investigation question.
The learner should connect:
changed root condition → root function → observed wilting → evidence-based explanation.
A strong answer uses both content knowledge and experimental structure.
What Should Be Kept the Same?
Do not memorise a giant universal list. The correct controls depend on the question.
Ask:
Which other condition could also change the result if I allowed it to vary?
For a heat investigation, cup size, water amount, starting temperature, time and surroundings may matter.
For a shadow investigation, the object, source, screen and relevant distances may matter.
For a plant investigation, plant type, plant size, water, soil, light and duration may matter.
A control is useful because it removes or reduces alternative explanations.
Changing Two Things at Once
Suppose a pupil compares:
- a small metal spoon in hot water;
- a large wooden spoon in warm water.
If the handles end at different temperatures, what caused the difference?
Possibilities include material, size, starting water temperature or several factors together.
The experiment is poor for isolating one cause because multiple important conditions changed.
Measurement Is Part of the Evidence
A measurement is not just a number. It includes:
- what property was measured;
- the value;
- the unit;
- the apparatus or method when relevant;
- the time or condition at which the reading was taken.
“The temperature was 45°C after 10 minutes” carries more scientific information than “It was 45.”
Mass, Volume and Temperature
Primary 4 commonly uses measurements such as:
- mass with a suitable balance or scale;
- liquid volume with graduated apparatus;
- temperature with a thermometer.
The learner should not only know the instrument. They should understand why that measurement answers the investigation question.
Read the Scale, Then Read the Meaning
A pupil can read a thermometer correctly and still answer the Science question wrongly.
If Set-up A changes from 60°C to 45°C and Set-up B changes from 70°C to 50°C, then:
- A decreased by 15°C;
- B decreased by 20°C.
Set-up B has the higher final temperature but the greater temperature decrease.
Measurement reading is only the first layer. Interpretation is the second.
Tables: Read Headings Before Numbers
Before comparing values, ask:
- What does each column represent?
- What are the units?
- What condition does each row represent?
- Am I comparing final values, differences, totals or changes?
Many “careless” errors are actually table-structure errors.
Original Table Case: Mass and Volume
| Object | Mass | Volume |
|---|---|---|
| P | 120 g | 80 cm³ |
| Q | 100 g | 120 cm³ |
| R | 150 g | 100 cm³ |
Question A: Which object has the greatest mass? R.
Question B: Which object occupies the greatest volume? Q.
Learning point: the “largest” object depends on what property is being compared. Mass and volume are different.
Diagrams: Position Can Carry Meaning
A diagram can show:
- parts;
- connections;
- directions;
- distances;
- sequence;
- before-and-after states;
- which condition differs between set-ups.
Read all labels and arrows. Do not assume two objects are connected merely because they are drawn close to each other.
Graphs: Ask What the Axes Mean
When a school introduces simple line or bar graphs, the same evidence rules apply.
Check:
- horizontal-axis variable;
- vertical-axis variable;
- units;
- overall pattern;
- specific values only when needed.
A rising line means the vertical quantity increases as the horizontal quantity changes over the displayed interval. It does not automatically prove the cause unless the investigation design supports that relationship.
Observation Is Not Inference
Observation: “The leaf drooped after two days.”
Inference: “The plant may not have absorbed enough water.”
Explanation: “The plant had many damaged roots. Roots absorb water, so the damaged root system reduced its ability to absorb sufficient water, causing more severe wilting.”
These statements perform different jobs.
Prediction Needs a Relationship
A prediction should not be a lucky guess.
Suppose the data show:
| Distance from torch | Shadow width |
|---|---|
| 10 cm | 18 cm |
| 20 cm | 13 cm |
| 30 cm | 10 cm |
A pupil might predict that moving the same object somewhat farther from the torch, while keeping the rest of the tested arrangement unchanged, would produce a smaller shadow than at 30 cm.
The prediction uses the observed pattern. The pupil should still recognise that a prediction beyond measured data has more uncertainty than a direct observation.
Conclusions Need Boundaries
Compare these:
Too weak: “Cup B is better.”
Useful: “Cup B’s water cooled less over 20 minutes.”
Evidence-based: “Under the tested conditions, the wrapping around Cup B reduced the temperature decrease compared with Cup A.”
Too broad: “This material is always the world’s best insulator.”
The strongest answer is not the longest. It is the one whose claim matches the evidence.
Can One Result Be Trusted?
In some classroom investigations, a single reading is enough to demonstrate a simple relationship. In others, repeating measurements can reduce the chance that one unusual reading dominates the conclusion.
Primary 4 learners can develop the practical habit of asking:
- Would repeating the measurement help?
- Were readings taken consistently?
- Does one result look very different from the others?
- Could the method itself create variation?
This is the beginning of evidence quality, without requiring advanced statistical treatment.
Unexpected Results Are Information
Suppose three trials give shadow widths of 12 cm, 12 cm and 25 cm under supposedly identical conditions.
Do not immediately erase 25 cm because it is inconvenient.
Ask:
- Was the ruler read correctly?
- Did the object move?
- Did the torch position change?
- Was the shadow edge difficult to define?
- Should the trial be repeated?
An unexpected result can expose a method problem or a hidden condition.
Procedure, Observation and Measurement
These are also different:
- Procedure: “Place the cup 20 cm from the lamp.”
- Observation: “A shadow forms on the screen.”
- Measurement: “The shadow is 14 cm wide.”
When a question asks what the pupil should do, give a procedure. When it asks what the pupil observes, give an observation. When it asks what to measure, name the property and method.
How to Improve a Method
A method improvement should solve a specific weakness.
| Weakness | Possible improvement | Why it helps |
|---|---|---|
| Different water amounts | Measure the same volume into each cup | Reduces an alternative cause of temperature differences |
| Different starting temperatures | Begin at the same temperature | Makes cooling comparison more meaningful |
| Object shifts during shadow measurement | Mark the object position | Keeps distance consistent |
| Readings taken at different times | Use the same timing schedule | Makes the comparison refer to equal durations |
| One doubtful measurement | Repeat the trial | Checks whether the result is repeatable |
“Do the experiment more carefully” is usually too vague because it does not identify what should change.
The Evidence-Based Answer Frame
For many explanation questions, this frame is useful:
EVIDENCE → SCIENTIFIC IDEA → LINK → CONCLUSION
Example:
Evidence: Cup B’s temperature decreased by only 12°C compared with 20°C for Cup A.
Scientific idea: Poor conductors reduce heat transfer compared with good conductors.
Link: The wrapping reduced heat transfer from the hot water to the cooler surroundings.
Conclusion: Therefore the water in Cup B stayed warmer.
This frame should be used flexibly. Not every question requires four separate sentences.
Command Words: Match the Answer to the Job
- State: direct answer.
- Describe: what happened or what pattern is shown.
- Explain: why it happened using the relevant scientific relationship.
- Compare: identify similarities or differences using the same property.
- Predict: extend a supported relationship cautiously.
- Suggest: propose a reasonable answer using the evidence and science.
A correct scientific fact can still be a poor response if it does not perform the requested job.
Original Investigation Workshop 1: Which Cover?
Three identical cups each contain 100 mL of water at 70°C.
| Cup | Cover | Temperature after 15 min |
|---|---|---|
| P | No cover | 52°C |
| Q | Card cover | 57°C |
| R | Foam cover | 61°C |
Q1. What factor was changed?
Q2. What was measured?
Q3. Name two important conditions that should be kept the same.
Q4. Which cup had the smallest temperature decrease?
Q5. Give a bounded conclusion.
Workshop 1 Answers
1. The cover condition/material.
2. Water temperature after the same duration, or temperature change if calculated.
3. Any two relevant controls such as cup type, water volume, starting temperature, time and surroundings.
4. Cup R. It fell from 70°C to 61°C, a decrease of 9°C.
5. Under these tested conditions, the foam-covered cup had the smallest temperature decrease over 15 minutes.
Original Investigation Workshop 2: Shadow Shape
A pupil wants to investigate whether changing the shape of a card affects the shape of its shadow.
The pupil uses three cards: a circle, triangle and star. But the pupil also places each card at a different distance from the torch.
Q1. What is wrong with the method?
Q2. How should it be improved?
Q3. What should be observed or measured?
Workshop 2 Answers
1. More than one relevant factor changes: both card shape and distance from the torch. This makes the cause of any shadow difference unclear.
2. Keep the distance, torch, screen and other relevant conditions the same while changing only the card shape.
3. Observe and record the resulting shadow shape, using a consistent method.
Original Investigation Workshop 3: Air Occupies Space
An empty-looking cup is pushed upside down into a basin of water. Water does not fill the cup completely. When the cup is tilted, bubbles escape and water rises into the cup.
Q1. What is the key observation?
Q2. What conclusion does the observation support?
Q3. Why is “the cup was empty at first” scientifically misleading?
Workshop 3 Answers
1. Water initially does not fill the upside-down cup, and bubbles escape when it is tilted.
2. Air occupied space inside the cup.
3. The cup contained air. It looked empty of visible solids or liquids, but it was not empty of matter.
Original Investigation Workshop 4: Plant Condition
Four similar seedlings are grown for one week.
| Plant | Roots | Water per day | Light | Final observation |
|---|---|---|---|---|
| A | Healthy | 50 mL | Bright | Leaves firm |
| B | Many damaged | 50 mL | Bright | Leaves wilted |
| C | Healthy | 20 mL | Bright | Leaves slightly wilted |
| D | Healthy | 50 mL | Dim | Leaves firm |
Q1. Which pair is most useful for comparing the effect of root damage while other listed conditions stay the same?
Q2. What conclusion can that pair support?
Q3. Why should Plant C not be used as the only comparison for root damage?
Workshop 4 Answers
1. Plants A and B.
2. Under the listed conditions, the plant with many damaged roots wilted while the plant with healthy roots remained firm, supporting the idea that root damage reduced the plant’s ability to maintain adequate water uptake.
3. Plant C differs in water amount, not root condition, so it tests a different factor.
Error Analysis: Why Did the Answer Fail?
After marking, classify the failure.
| Error type | Example | Repair |
|---|---|---|
| Question-reading | Explains when asked to state | Underline command and target |
| Variable | Names measured result as changed factor | Reconstruct change/measure/keep roles |
| Measurement | Reads wrong scale or omits unit | Instrument practice |
| Data | Compares final values instead of changes | Calculate and label differences |
| Concept | Says cold flows into warm water | Repair heat-transfer model |
| Evidence | Makes universal claim from one set-up | Bound conclusion to tested conditions |
| Expression | Knows idea but leaves causal link unstated | Evidence → idea → link → conclusion |
| Transfer | Can answer textbook diagram only | Change surface example |
This prevents every mistake from being called “careless”. Different failures require different repairs.
The Correction Book Should Store Decisions, Not Just Answers
Instead of copying the model answer alone, record:
- what the question asked;
- what decision was missed;
- the scientific relationship needed;
- the corrected reasoning;
- a new transfer example.
For example:
Error: compared final temperatures instead of temperature decreases.
Repair rule: if the question asks “which changed more?”, calculate or compare the change from each starting value.
Transfer: repeat with mass gain, shadow-size difference or volume change.
Open Book → Reduced Help → Closed Book
Strong independence is built progressively.
Stage 1: solve with notes and teacher prompts.
Stage 2: solve with the QUESTION → CHANGE → KEEP → MEASURE → READ → CONCLUDE routine visible but without worked answers.
Stage 3: solve without prompts.
Stage 4: solve an unfamiliar question after a delay.
If the learner only succeeds at Stage 1, the knowledge is not yet independently retrievable.
Transfer Is the Final Test
After learning heat with cups, use metal spoons or insulated lunch containers.
After learning shadow variables with a toy, use a card cut-out.
After learning liquid volume with water, use coloured liquid in differently shaped vessels.
After learning plant-root effects, redraw the experiment with a different plant and different observation wording.
Keep the underlying relationship. Change the surface.
How to Know Whether the Learner Really Understands
Ask the pupil to do four things without notes:
- Explain the scientific idea in simple language.
- Recognise it in a new set-up.
- Identify the evidence that matters.
- Predict what should change if one condition changes.
A learner who can only repeat the definition has memory. A learner who can do all four has a more usable model.
A 30-Minute Investigation Lesson
Minutes 1–5: retrieve one concept.
Minutes 6–10: read an investigation and label changed, measured and controlled conditions.
Minutes 11–15: interpret one table or diagram.
Minutes 16–20: write a bounded conclusion.
Minutes 21–25: identify one method limitation and propose a specific improvement.
Minutes 26–30: change the surface example and repeat independently.
This is an eduKate teaching suggestion, not an official programme.
Parent and Tutor Diagnostic Questions
- “What exactly did the pupil change?”
- “What was measured?”
- “Which other condition could also have changed the result?”
- “What evidence in the table supports your conclusion?”
- “Are you comparing a final value or a change?”
- “Does your conclusion go further than the data?”
- “What would you repeat or improve?”
- “Can you solve the same relationship with different objects?”
If the child cannot answer these questions, more worksheets may simply repeat the same hidden weakness.
Primary 4 Examination Craft Without Over-Teaching the Exam
Primary 4 assessment differs across schools. The safest preparation is not to predict one exact paper format. Build capabilities that survive different formats:
- read instructions carefully;
- interpret representations;
- select relevant evidence;
- use correct scientific relationships;
- write concise causal explanations;
- check units and comparisons;
- manage unfamiliar set-ups without panic.
These skills support later PSLE Science without turning Primary 4 into premature PSLE drilling.
The Primary 5 Bridge
Primary 5 Science increases systems complexity and variable interactions. Investigation skills become even more valuable because learners must coordinate content, representation, evidence and explanation at the same time.
A Primary 4 learner who can reliably identify what changed, what was measured and why the evidence supports a conclusion already has a strong part of the future PSLE reasoning engine.
Complete Batch 1 | Primary 4 Science Learning Guide
- Primary 4 Science Learning Guide | Relationships, Evidence and Explanation
- Primary 4 Science Learning Guide | Plant and Human Systems
- Primary 4 Science Learning Guide | Matter, Light and Heat
- Primary 4 Science Learning Guide | Investigations, Data, Answers and Transfer
Return to the Primary 4 Science Learning Hub or the wider Primary Science | Complete P1–P6 and PSLE Science Guide.
The Quiet Return
An investigation becomes manageable when the learner stops seeing a wall of apparatus and numbers.
Find the question. Find the change. Find what was kept comparable. Read the measurement. Match the evidence to the scientific idea. Make only the conclusion the evidence can carry. Then change the surface and see whether the reasoning still works.
