How to Track a Changing Comparison Reference Across a Multi-Part PSLE Science Question

Wait, What? The Science Did Not Change — the Comparison Did.

A question shows three set-ups: P, Q and R. Part (a) asks you to compare P with Q. You do that correctly. Part (b) then asks how R differs from Q. You keep thinking about P because P was the reference in the first part. Your Science knowledge is fine. Your reading is fine. Your mistake is smaller and more dangerous: the comparison reference changed, but your reasoning did not reset.

Multi-part PSLE Science questions can reuse the same diagram, table, experiment or system while asking a different scientific job in each part. Earlier information may still matter, but an earlier comparison does not automatically remain the comparison for the next sub-question.

Quick Answer

For every new sub-question, rebuild the comparison before you answer. Identify what quantity or property is being compared, which object or set-up is the target, which object or set-up is the reference, under which condition, and at which time or stage. Keep earlier evidence that is still relevant, but do not carry an old reference forward just because it appeared first.

A useful thinking line is:

NEW PART → NEW JOB → NEW TARGET → NEW REFERENCE → SAME OR NEW CONDITION? → SAME OR NEW TIME? → THEN COMPARE.

Owned PSLE Science Learning Job

This guide owns one narrow learner job: tracking and resetting the comparison reference across a multi-part PSLE Science question. It does not own the scientific concepts used in the examples. Those concepts remain with their existing Science pages. The goal here is to keep your comparison logic attached to the exact sub-question you are answering.

Why This Matters in the Current PSLE Science Frame

For examination from 2026, Standard PSLE Science assesses the 2023 Primary Science syllabus. The official assessment objectives include knowledge with understanding, application of scientific facts, concepts and principles, and scientific inquiry such as interpreting and analysing information, evaluating observations, information and methods, and communicating explanations and reasoning.

A changing comparison reference sits inside those larger abilities. The question may not announce, “Your reference has changed.” You have to read the relationship represented by the information and direct your Science toward the current task.

The Comparison Address

Before you compare, give the comparison an address. Five pieces are usually enough:

Quantity or property: What exactly is being compared?
Target: Which object, organism, set-up, stage or result is the question asking about?
Reference: Compared with what?
Condition: Under what stated circumstances?
Time or stage: At the start, after the same duration, at a stated point, or at the end?

If one of these changes between part (a) and part (b), the comparison has changed even when the diagram has not.

Part	Quantity	Target	Reference	Condition / time
(a)	temperature after 10 min	Set-up P	Set-up Q	same starting temperature, after 10 min
(b)	temperature after 10 min	Set-up R	Set-up Q	same starting temperature, after 10 min
(c)	change in temperature	Set-up R	its own starting value	from 0 to 10 min

Notice what happened. The information source stayed the same. The scientific quantity changed in part (c), and the reference changed from another set-up to the same set-up’s own starting value.

The PSLE Science Reasoning Chain

Use the series reasoning law, but place a comparison reset at the front:

READ THE CURRENT SUB-QUESTION. Do not answer from memory of the previous part.
IDENTIFY THE SCIENTIFIC OBJECT OR RELATIONSHIP. What is the target quantity, object or relationship now?
NAME THE REFERENCE. Compared with what, exactly?
DISTINGUISH OBSERVATION FROM INFERENCE. What is directly shown, and what must be reasoned?
SELECT THE RELEVANT CONCEPT. Use only the concept needed for this comparison.
EXPLAIN THE CAUSAL MECHANISM if the command requires explanation.
CONNECT TO THE QUESTION’S CONDITION. Keep time, set-up and stated conditions aligned.
STATE THE OUTCOME. Make the comparison explicit.
CHECK AGAINST THE EVIDENCE. Confirm that your target and reference are the ones named in this part.

Worked Reasoning Example 1: Same Table, New Reference

Imagine three identical containers begin with water at the same temperature. They are wrapped with different numbers of identical insulating layers. After ten minutes, the temperatures are recorded.

Set-up	Number of layers	Temperature after 10 min / °C
P	1	54
Q	2	59
R	4	64

Part (a): Compare P with Q. Your reference can be Q. P has a lower temperature after ten minutes than Q.

Part (b): Compare R with Q. The reference is still Q, but the target changes from P to R. R has a higher temperature after ten minutes than Q.

Part (c): State how much the temperature of R changed from the start if its starting temperature was 70°C. The comparison is no longer R versus Q. It is R at the end versus R at the start. The relevant values are 64°C and 70°C. The decrease is 6°C.

The trap is not mathematical. The trap is comparison ownership. Part (c) asks for a within-set-up change, not a between-set-up difference.

Worked Reasoning Example 2: The Reference Changes but Earlier Evidence Still Matters

Suppose a diagram shows three plants, X, Y and Z. X and Y receive the same amount of water but different light conditions. Y and Z receive the same light condition but different amounts of water. The question has two parts.

Part (a) asks you to compare X and Y. That comparison isolates one difference in the stated set-up. Part (b) asks you to use Y and Z. The shared reference set-up is now Y for a different comparison. Your knowledge from part (a) may still help you understand the diagram, but you must not use the X–Y relationship as though it were the Y–Z relationship.

This is why a good learner does not erase earlier reasoning. Instead, the learner stores it with a label: “X versus Y under these conditions.” When the next part begins, the learner creates a new label.

What Changes and What Stays the Same?

Multi-part questions are difficult because some information continues while other information resets. Treat the question like a scientific state that must be updated carefully.

May stay the same	May change
diagram, table or experimental set-up	target object
scientific concept	comparison reference
some earlier observations	quantity being asked about
background conditions	time point or stage
labels of set-ups	command word and answer job

The learner’s job is not to restart the whole question from zero. It is to decide what to carry and what to reset.

Observable Failure Signatures

You write a scientifically true comparison, but it is between the wrong two set-ups.
You use the same reference word in every sub-question even though the prompt changes it.
You calculate change from another set-up instead of from the target’s own starting value.
You answer part (b) using the conclusion from part (a) without checking whether the conditions match.
You copy an earlier sentence and change only one label.
You say “more” or “less” without being able to state “more than what?”
You compare values from different time points because one was used in the previous part.

Find the Earliest Weak Link

Do not diagnose every wrong multi-part answer as carelessness. Ask where the first mismatch occurred.

If the wrong set-ups were chosen before any Science was used, the weak link is reference selection.
If the correct set-ups were chosen but the wrong time points were compared, the weak link is time alignment.
If the target and reference were correct but the answer uses the wrong quantity, the weak link is quantity identity.
If all comparison parts are correct but the reason is wrong, the weak link is concept or mechanism.
If the reasoning is correct aloud but labels are swapped in writing, the weak link is representation control.

Repair the earliest weak link. Rewriting the entire chapter will not fix a reference-selection error.

A 20-Second Comparison Reset Protocol

Box the current sub-question number.
Underline the quantity or property being asked about.
Circle the target set-up or object.
Draw a small arrow from the target to the reference: target → compared with reference.
Write the time or condition beside the arrow if it matters.
Only then read the values or choose the concept.

This is a practice protocol, not an official examination notation and not a compulsory format. Its purpose is to make the reference visible while you are training.

Do Not Confuse Comparison With Explanation

First establish the correct comparison. Then decide whether the question asks you to state, describe, compare or explain.

If the task is only to compare, a clear evidence-based difference may be enough for that job. If the task asks you to explain why the difference occurs, you must then select the relevant scientific concept and connect the changed condition to a mechanism and outcome. A perfect mechanism attached to the wrong comparison is still the wrong answer to the current task.

Common Traps

The first set-up becomes the permanent reference. It should not. The wording of each sub-question decides the reference.
The largest value becomes the reference. Size does not choose the comparison; the question does.
The control set-up becomes the reference for every part. A control may be scientifically important, but a later part may ask for a different comparison.
The previous answer becomes new evidence. Your earlier conclusion is not automatically a measured observation. Keep evidence and inference separate.
“More” is treated as complete. Always know the scientific quantity and the reference.
Same diagram means same task. One representation can support several different learner jobs.

Retrieval and Practice Sequence

Train this skill separately before burying it inside full papers.

Reference spotting: Take five short comparisons and name only target and reference. Do not solve the Science yet.
Reference switching: Use one table and write three different comparison prompts that change the reference.
Quantity switching: Compare final value in one part, amount of change in another, and time taken in a third.
Condition switching: Keep the same objects but change the stated time or condition.
Full reasoning: Add the scientific mechanism only after the comparison is stable.
Mixed return: Two days later, solve a new multi-part example without the protocol visible.

Unfamiliar Transfer Test

You understand this skill when you can handle a completely different surface example — a plant investigation, a heat graph, a force diagram, a materials table or a system question — and still identify the target and reference independently. The scientific topic may change. The comparison job should survive.

Delayed Independent Return Test

After a delay, return to one multi-part question with your notes closed. Before answering each part, write a tiny comparison address: quantity, target, reference, condition, time. If you can do that without prompts and your final answer uses the correct evidence, the skill is becoming independent. If you still carry an old reference forward, return to reference-switching practice rather than doing another whole paper.

Answer and Checking Receipt

Before leaving a multi-part item, ask four questions:

Did I answer the current part, not the previous part?
Can I name my target and reference without looking at my sentence?
Are the compared values from the same quantity, condition and required time?
If I explained a cause, does my explanation actually account for this comparison?

Parent and Tutor Teaching Guide

When a child gets a later sub-question wrong, do not immediately reteach the concept. Ask the learner to point to the two things being compared. Then ask, “Compared with what in this part?” If the learner names the previous part’s reference, the problem has already been located.

Use one representation and deliberately ask several different questions about it. This is more diagnostic than giving five unrelated worksheets. For example, one table can support a between-set-up comparison, a within-set-up change, a final-value question and a time-aligned comparison. The child learns that the evidence can stay while the job changes.

Fade the annotations gradually. First allow target/reference arrows. Then ask for them mentally. Finally test a fresh multi-part question after a delay. The goal is not permanent annotation; it is stable internal control of the comparison.

Useful Internal Routes

Authoritative References

These official sources define the curriculum and assessment frame. The comparison-reset protocol in this guide is an eduKate learning tool, not an official marking formula or compulsory examination method.

The Quiet Return

A multi-part Science question is not one long answer. It is a sequence of small scientific jobs sharing a common information source.

Carry the evidence that still belongs. Carry the concept that still belongs. Carry the object identity that still belongs.

But when the question changes the comparison, let the old reference go.

Small Group Tutorials