Wait, What? Three Set-ups Do Not Mean You Should Compare All Three at Once
A PSLE Science question can show three, four or even more experimental set-ups. That often makes the page look difficult before the Science has even started. The learner sees many labels, several measurements and a table full of results, then tries to explain everything in one sentence.
But the important scientific job is usually smaller: choose the comparison that can answer the exact question.
If the question asks about moving air, you need set-ups that differ in moving air while the other relevant conditions are kept the same. If it asks about exposed surface area, you need a different pair. The presence of four set-ups does not mean all four belong in every comparison.
When there are many set-ups, do not compare more. Compare more precisely.
Quick Answer
When a PSLE Science question contains three or more set-ups, first identify the factor the question wants you to investigate. Then find the pair of set-ups in which that factor changes while the other relevant conditions are sufficiently comparable. Read the measured outcome for that pair, describe the evidence, explain the relevant scientific mechanism, and state only the conclusion the comparison can support.
This page does not replace the canonical Primary Science guides on variables, fair tests, measurement or data tables. Those pages own the underlying scientific inquiry concepts. This guide owns a narrower learner job: how to select and use a valid comparison when several set-ups are presented together in a PSLE Science question.
The Official PSLE Science Frame
For examination from 2026, the PSLE Science Paper assesses attainment in the 2023 Primary Science syllabus. SEAB states that candidates are expected to demonstrate knowledge with understanding and to apply scientific knowledge and scientific inquiry. This includes making predictions or hypotheses, interpreting and analysing information, evaluating observations, information and methods, and communicating explanations and reasoning.
That matters here. Choosing a valid comparison is not an exam trick. It is part of scientific inquiry: deciding what the evidence can actually tell you.
Owned PSLE Science Learning Job
- Read a multi-set-up question without treating every set-up as equally relevant.
- Identify the factor or relationship the question is asking about.
- Select the pair or subset of set-ups that forms the cleanest comparison for that job.
- Check which important conditions are the same and which are different.
- Separate the observed result from the explanation for the result.
- Avoid attributing an outcome to one factor when several important factors changed.
- Use tables, diagrams and graphs as evidence rather than decoration.
- State a conclusion that is no broader than the comparison permits.
The PSLE Science Comparison Chain
READ THE EXACT QUESTION → IDENTIFY THE TARGET FACTOR OR RELATIONSHIP → SCAN ALL SET-UPS → FIND THE CLEANEST COMPARISON → CHECK WHAT IS THE SAME → CHECK WHAT IS DIFFERENT → READ THE OUTCOME → DISTINGUISH OBSERVATION FROM INFERENCE → APPLY THE RELEVANT SCIENCE → STATE THE CONCLUSION → CHECK THAT THE EVIDENCE REALLY SUPPORTS IT.
This is the same PSLE Science reasoning law used across the series, but the key move here is comparison selection.
Why Three or More Set-ups Feel Harder Than Two
With two set-ups, the comparison is often obvious. With four set-ups, several possible pairs exist. The learner has to decide which pair answers which question.
Four set-ups create six possible pairs. Five set-ups create ten possible pairs. You do not need to calculate these numbers during the exam. The point is that the page contains possible comparisons, not one giant comparison.
The safest move is to let the question choose the comparison.
A Worked Comparison Matrix
Imagine an original investigation about how quickly equal amounts of water evaporate. Four set-ups are prepared:
| Set-up | Container opening | Moving air | Temperature | Water lost after the same time |
|---|---|---|---|---|
| A | Wide | No | 30°C | 12 g |
| B | Wide | Yes | 30°C | 20 g |
| C | Narrow | Yes | 30°C | 11 g |
| D | Wide | Yes | 40°C | 28 g |
The Science concepts in this example are not the owner of this page. The learning job is how to choose comparisons.
If the question asks about moving air
Compare A and B. The container opening and temperature are the same; moving air differs. The measured outcome differs too. This pair can support reasoning about the effect of moving air under those conditions.
If the question asks about exposed surface area
Compare B and C. Moving air and temperature are the same; container opening differs. The pair is useful for the surface-area question.
If the question asks about temperature
Compare B and D. Container opening and moving-air condition are the same; temperature differs.
Notice what happened: the same set-up, B, can participate in several different valid comparisons. A set-up is not “the control” or “the experiment” forever. Its role depends on the question and the comparison being made.
The One-Difference Scan
For a simple causal comparison, ask:
- What factor does the question want me to examine?
- Which two set-ups differ in that factor?
- Which other important conditions are the same?
- Is the measured outcome comparable?
- Did anything else change that could also explain the result?
This is not a universal law that every real scientific investigation can change only one thing. Real research can use complex designs and statistical methods. At Primary level, however, the clean-comparison model is a powerful way to reason about simple investigations and fair tests.
Do Not Start With the Results Column
A common error is to look for the largest number first. Suppose D has the greatest water loss. A learner may immediately say, “D proves moving air causes the fastest evaporation.” But D differs from A in both moving air and temperature. That pair does not isolate one of those factors.
Large differences can attract attention. Scientific comparison requires something stricter: which conditions produced the difference, and can this comparison separate them?
Observation Is Not Yet the Explanation
From A and B, you may observe that B lost more water after the same time. That is evidence. The explanation requires the relevant scientific concept and mechanism.
Keep the order:
- Observation: B lost 20 g while A lost 12 g.
- Relationship: the relevant planned difference is moving air.
- Concept: choose the syllabus concept that explains the process.
- Mechanism: explain why the changed condition affects the process.
- Outcome: connect that mechanism back to the measured result.
Do not skip from “B is larger” to a memorised conclusion without showing why the comparison is meaningful.
When No Pair Gives a Clean Comparison
Sometimes the correct scientific judgement is that the information does not allow a clean conclusion about one factor.
Imagine A and D are the only set-ups shown. A has no moving air at 30°C. D has moving air at 40°C. If D loses more water, the difference may involve moving air, temperature, or both. A careful learner should not pretend the evidence separates them.
Scientific reasoning is not weakened by saying “this comparison cannot isolate the factor”. It is strengthened by refusing to claim more than the evidence supports.
The Pairwise Comparison Table
When the page feels crowded, make a tiny scratch table:
| Question asks about | Set-up 1 | Set-up 2 | Target difference | Other key conditions same? |
|---|---|---|---|---|
| Moving air | A | B | Air movement | Yes |
| Opening size | B | C | Opening | Yes |
| Temperature | B | D | Temperature | Yes |
This representation reduces working-memory load without turning the problem into a trick. It makes your scientific decision visible.
How to Read a Multi-Set-up Diagram
- Read the caption and labels before interpreting the picture.
- Identify what is identical across all set-ups.
- Mark what changes from one set-up to another.
- Notice whether the measured quantity is the same type in every set-up.
- Do not infer hidden properties from drawing size unless the question states that the drawing is to scale.
- Use the question wording to decide which differences matter.
How to Read a Multi-Set-up Results Table
Read the headings, units and conditions before comparing numbers. “20” is not meaningful until you know whether it means grams lost, seconds taken, centimetres grown or number of organisms observed.
- What does each row represent?
- What does each column represent?
- Are units the same?
- Was measurement time the same?
- Which rows form the comparison the question asks for?
- Is the pattern consistent enough to support the conclusion?
What If There Are Repeated Results?
If a table gives repeated trials, do not collapse them carelessly into one convenient number. Read what the question asks you to do. Repeated observations can help reveal consistency and variation, but the meaning of the repeats belongs to the inquiry design and evidence, not to a memorised “always take the average” rule.
If a calculation is required, follow the stated data and syllabus-appropriate method. If no calculation is required, you may still use the repeated results to judge whether a pattern is stable.
A Second Worked Example — Three Plants, Two Questions
Three similar young plants are observed for the same duration. Plant P receives light and adequate water. Plant Q receives no light and adequate water. Plant R receives light but much less water. The measured outcome is change in mass.
If the question asks about light, P and Q are the relevant comparison because water condition is held alike while light differs. If the question asks about water availability, P and R are more useful because light condition is alike while water differs.
Do not compare Q and R to claim the effect of light alone: both light and water conditions differ. Again, the skill is not knowing a plant fact. The skill is selecting the comparison that can test the proposed relationship.
A Third Worked Example — When an Option Uses the Wrong Pair
In a multiple-choice question, one option may quote two correct numbers from the table but compare the wrong set-ups. This is why familiar words and true data are not enough.
Test the option with three checks:
- Does it use the set-ups relevant to the factor being discussed?
- Does that pair control the other important conditions?
- Does the conclusion match the direction of the data?
A statement can contain true numbers and still be a scientifically invalid explanation.
Common Trap: Calling One Set-up “The Control” Without Asking Control for What?
Students sometimes memorise that Set-up A is “the control set-up” and treat that label as permanent. In a multi-comparison design, one set-up may serve as a reference for several comparisons, while another pair may answer a different question more cleanly.
Use the term only when you understand its role. The stronger question is: what comparison is this set-up helping me make?
Common Trap: Comparing Across Different Measurement Times
If one result was measured after 10 minutes and another after 30 minutes, the raw values may not be directly comparable for the question you want to answer. Check the measurement condition before declaring one process “faster” or “greater”.
Common Trap: Treating Similar-Looking Set-ups as Identical
Read labels. Two drawings can look the same but differ in material, temperature, amount, distance, time or another stated condition. Conversely, drawings may look different only because of illustration style while the stated scientific conditions are identical.
Earliest Weak-Link Diagnosis
If multi-set-up questions repeatedly go wrong, diagnose the first failed step rather than calling the whole topic weak.
| Failure signature | Likely earliest weak link | Repair |
|---|---|---|
| Chooses largest/smallest result immediately | Question target not identified | Cover the results column and decide the required factor first |
| Compares two set-ups with several differences | Conditions not mapped | List SAME and DIFFERENT before reading outcome |
| Finds correct pair but gives only numbers | Observation not connected to concept | Add mechanism and condition |
| Gives a broad “always” conclusion | Evidence boundary lost | Restate what the comparison actually tested |
| Uses every set-up in every answer | Cannot select relevant evidence | Ask “Which pair answers this exact question?” |
Misconception Repair: More Data Does Not Automatically Mean Better Evidence
More rows, more measurements and more set-ups can help, but only if they answer the scientific question in a valid way. Ten poorly matched comparisons do not become strong evidence by quantity alone.
At PSLE level, precision in comparison is usually more useful than trying to use everything on the page.
Misconception Repair: “Only One Thing Changed” Needs Context
In a simple fair test, learners often aim to change one planned factor while keeping other relevant conditions constant. That is a useful Primary Science model for investigating cause and effect.
But real-world investigations can be more complicated. This guide teaches the syllabus-appropriate reasoning needed to judge simple comparisons; it does not claim that all scientific research uses only two perfectly identical groups.
Question-Reading Protocol for Multi-Set-up Problems
- 1. Read the command. Is the task to compare, explain, predict, evaluate or identify?
- 2. Name the target. What factor or relationship is the question about?
- 3. Map the set-ups. What changes? What stays the same?
- 4. Select the comparison. Which pair gives the cleanest evidence?
- 5. Read the result. What was actually observed or measured?
- 6. Explain. Which scientific concept and mechanism connect condition to outcome?
- 7. Limit the claim. What does the evidence support—and what does it not support?
Retrieval Practice Sequence
Do not revise this skill by reading the guide repeatedly. Retrieve the decisions.
- Day 1: Given four set-ups, identify the best pair for three different target factors.
- Day 1: Explain why one tempting pair is invalid for a causal conclusion.
- Day 2: Use a table instead of a diagram.
- Day 4: Use a graph plus setup descriptions.
- Day 7: Solve an unfamiliar investigation without being told which pair to compare.
- Later: Explain the decision to another person without notes.
Unfamiliar Transfer Test
You are given four unfamiliar boxes containing a material. The boxes differ in one or more conditions: temperature, material thickness and whether air is moving. The measured outcome is cooling time. You have never seen the experiment before.
Can you still identify which pair tests the effect of thickness? If yes, the skill has transferred. You did not need to recognise the exact experiment. You used the structure of scientific comparison.
Delayed Independent Return Test
Two or three days later, take a new four-set-up investigation. Without hints, can you:
- state the exact target factor;
- choose the comparison;
- identify the conditions that must match;
- read the evidence correctly;
- explain the mechanism;
- reject an invalid comparison;
- state the limit of the conclusion?
If you can do this on a changed context, the learning is becoming independent rather than tied to one worksheet.
Answer-Checking Receipt
- Did I compare the set-ups that answer the question?
- Did I check the other relevant conditions?
- Did I distinguish observation from inference?
- Did I use the measured outcome correctly?
- Did I connect the condition to the mechanism?
- Did I state the outcome in the correct direction?
- Did I avoid claiming a cause when several important factors changed?
- Did I keep the conclusion within the evidence?
Parent and Tutor Teaching Guide
When a child is overwhelmed by four set-ups, do not immediately tell them which pair to use. Ask one narrowing question: “What factor is this question asking about?”
Then ask the child to circle only two set-ups that can investigate that factor. Follow with:
- What is the same?
- What is different?
- What was measured?
- What did you observe?
- What scientific idea explains the observation?
- What can you safely conclude?
If the learner chooses the wrong pair, resist giving the final answer. Ask them to list the differences between the chosen set-ups. Often the invalid comparison becomes visible immediately.
Useful eduKate Routes
- How to Decode Variables and Fair Tests in PSLE Science Questions
- How to Turn PSLE Science Diagrams, Tables and Graphs Into Evidence for an Answer
- How to Evaluate a PSLE Science Experiment and Improve the Method
- How to Identify What Evidence a PSLE Science Question Actually Gives You
Authoritative External References
- Singapore Examinations and Assessment Board — PSLE Science syllabus for examination from 2026.
- Singapore Ministry of Education — Science Teaching & Learning Syllabus, Primary, 2023.
- Education Endowment Foundation — 2023 systematic review of approaches to Primary Science teaching, used as teaching evidence rather than as a Singapore examination rule.
The Quiet Ending
A crowded investigation does not require a crowded answer. The learner’s job is to find the comparison that can speak clearly.
Read the question. Choose the pair. Check what stayed the same. Notice what changed. Read the result. Explain the mechanism. Then stop where the evidence stops.
That is not merely a way to survive a multi-set-up PSLE Science question. It is one of the central habits of scientific reasoning: compare only what can be compared, and claim only what the comparison can support.
