How to Read a PSLE Science Result After Two Samples Have Been Combined

Wait, what? Once two separately tracked samples are combined, a later result may no longer belong to either original sample on its own.

This sounds obvious when stated slowly. It becomes much harder inside a multi-stage PSLE Science question. Learners often remember that Sample A had one earlier value and Sample B had another, then unconsciously assign a later measurement of the combined material back to A or B. The number is real. The measurement may be accurate. The ownership of the evidence is wrong.

Quick Answer

Before combination, Sample A and Sample B have separate identities and separate evidence histories. At the combination event, ask whether the question still preserves a way to distinguish their contributions. If not, later observations or measurements usually describe the combined system, not A alone and not B alone. Keep the earlier evidence about A and B, but do not push a combined result backward into either original sample unless the method gives evidence that allows you to do so.

The PSLE Science Learning Job This Guide Owns

This guide owns one reasoning job: preserving evidence provenance when separately tracked samples, portions or groups are merged before a later observation. It is not a concept page about mixtures, solutions, temperature, materials or living things. Those scientific concepts remain owned elsewhere. Here, the learner is practising identity, evidence and inference through a merge event.

The reasoning chain is:

READ the evidence for each separate sample.
IDENTIFY the moment they are combined.
ASK what identity can still be tracked after that moment.
ATTACH later observations to the unit that was actually observed or measured.
DISTINGUISH a combined-system result from an individual-sample result.
SELECT the relevant scientific concept only after evidence ownership is clear.
EXPLAIN the mechanism under the stated conditions.
CHECK that no combined result has been assigned backward without evidence.

Why This Is a Scientific Inquiry Skill

SEAB states that the 2026 PSLE Science paper assesses the 2023 Primary Science syllabus and includes interpreting and analysing information, evaluating observations, information and methods, and communicating explanations and reasoning. A learner who cannot tell what a measurement belongs to cannot interpret that measurement safely. Evidence needs an owner: the object, sample, group or combined system that was actually measured.

This guide does not claim that national examination answers require the labels used here. They are learning tools for making the evidence structure visible.

The Merge Boundary

Stage	Evidence ownership	Main question
Before combination	A and B can be tracked separately.	What do we know about A? What do we know about B?
Combination event	The system changes from two tracked units to a combined unit unless separation is preserved.	What exactly was combined, and was any identifying information retained?
After combination	New measurements belong to the combined system unless the method measures a distinguishable part.	What object or system was actually measured?
Conclusion	Claims must match the level of evidence.	Can this result support a statement about A, B, or only A+B?

The merge boundary is the moment when evidence ownership can change.

Worked Example 1: Two Liquid Samples Become One Measured System

Imagine an original practice question. Liquid Sample A has one measured temperature and Sample B has another. They are then poured into one container. Later, the temperature of the liquid in that container is measured.

Before pouring, there are separate measurements for A and B. After pouring, the later thermometer reading describes the liquid at the place and time measured in the combined container. It does not automatically mean “A now has this temperature” while B somehow retains a different hidden value.

Equally, you should not assume the combined liquid is perfectly uniform unless the question, method or relevant science supports that interpretation. Evidence ownership and evidence uniformity are different questions.

The safe sequence is:

A before combination: keep A’s stated measurement.
B before combination: keep B’s stated measurement.
Combination: mark a new combined system.
Later reading: attach it to the location and combined material actually measured.
Explanation: use the relevant scientific concept only after that evidence map is secure.

Worked Example 2: Two Groups Are Combined and the Labels Are Lost

Imagine two groups of similar specimens, Group P and Group Q, that experienced different earlier conditions. Before the groups are combined, each specimen is identifiable by group. Later, the specimens are placed together and their labels are removed. A final count is recorded for the combined group.

The final count may tell you something about the combined group. If the method no longer lets you identify which final specimens came from P and which came from Q, the final total cannot automatically be split between P and Q.

A learner might be tempted to use the earlier proportions to estimate each group’s contribution. That would be an additional inference. It would need evidence and assumptions that the question may not provide. The scientifically disciplined answer keeps the final evidence at the combined-group level unless traceability remains.

Worked Example 3: Combination Does Not Always Destroy Identity

Now change the method. Suppose P specimens remain marked with one symbol and Q specimens with another even after both groups are placed in the same container. Later observations are recorded separately for each symbol.

Here, physical combination does not necessarily destroy data identity. The samples share a setting, but the method preserves traceability. You may still attach later evidence to P and Q separately if the observations are actually recorded separately.

This is why the correct rule is not “combined means impossible to separate”. The correct rule is:

After a merge, assign evidence only to identities the method still allows you to track.

Observable Failure Signatures

You use a final combined measurement as if it were measured separately for A and B.
You forget the exact point where the samples were merged.
You erase useful pre-combination evidence because “they are one sample now”.
You split a later total between original groups without a method for identifying contributions.
You assume a combined sample is uniform everywhere without evidence.
You keep using old labels after the method says the identities can no longer be distinguished.
You do the opposite: ignore labels that actually preserve identity after combination.

Find the Earliest Weak Link

If a final conclusion is wrong, diagnose backwards:

Was the later observation read correctly?
Was it attached to the correct measured object or system?
Did the learner notice the merge event?
Did the method preserve or destroy separate traceability?
Only then: was the scientific mechanism correct?

If step 2 or 3 fails, the problem is evidence provenance. Re-teaching a content fact will not repair it.

The MERGE Protocol

M — Map the separate identities first

Write what is known about A and what is known about B before combination. Keep quantities, units, conditions and times attached to the correct sample.

E — Establish the merge point

Mark the exact procedure step where the separate units become one system or begin sharing the same space.

R — Retain only traceability the method preserves

Labels, coloured markers, partitions, separate sensors or recorded positions may preserve identity. If none exist, do not invent them.

G — Give each later result its correct owner

Ask what exactly was observed or measured: A, B, one marked subgroup, one location, or the combined system.

E — Explain only after ownership is secure

Then select the relevant scientific concept and build the causal explanation from conditions to mechanism to outcome.

A Data Table Can Reveal the Evidence Owner

Look closely at row and column headings. If the table changes from separate rows labelled A and B to one row labelled “combined sample”, that change matters. If later rows still record A and B separately, traceability may have been preserved.

Do not compare numbers before asking what one row represents. A perfectly read number attached to the wrong scientific unit is still bad evidence use.

Combination, Average and Total Are Different Ideas

A combined sample is a physical or procedural unit created from previously separate material or groups. An average is a calculated summary of several values. A total is an accumulated or counted quantity. These ideas can appear together, but they are not interchangeable.

Idea	What happened?	What evidence question follows?
Combined sample	Separate units were physically/procedurally brought together.	Can their identities still be tracked?
Average	Separate values were mathematically summarised.	Which individual values produced the summary?
Total	Amounts or counts were added or accumulated.	Does the total preserve information about individual contributions?

Do not infer one from another unless the question supports it.

Misconception Repair: “If I Know the Combined Result, I Know Both Original Results”

Usually, one combined result is not enough to reconstruct two unknown individual results. Many different pairs can produce the same total, average or combined observation depending on the quantity involved. The exact scientific relationship matters.

The replacement rule is:

Do not reverse a merge unless the evidence gives you a scientifically justified way to do so.

Practice Sequence

Identity practice: label A and B before combination.
Merge practice: mark the exact procedure step where evidence ownership can change.
Traceability practice: decide whether labels or method features preserve identity afterward.
Result practice: attach each later result to A, B, a subgroup or A+B.
Inference-limit practice: state one thing the combined result cannot tell you about the originals.
Transfer practice: repeat with a different Science theme and different type of data.
Delayed return: solve a fresh merge scenario several days later without the protocol visible.

Unfamiliar Transfer Challenge

Create an original case with Sample X and Sample Y. Give each one an earlier measurement. Combine them. Then give one later observation. Ask yourself:

What do I still know about X before the merge?
What do I still know about Y before the merge?
What does the later observation actually belong to?
Can I say what X alone did after the merge?
What extra method feature would be needed if I wanted to keep tracking X and Y separately?

Then redesign the case so X and Y retain labels after combination. Notice how the evidence you are allowed to claim changes even though the physical combination still occurs.

Delayed Independent Return Test

After a delay, take a fresh multi-stage question and draw a vertical line at every merge event. To the left, list separately owned evidence. To the right, list only evidence that the later method still lets you assign to a specific identity.

You have a strong receipt when you can do this mentally and your final answer no longer smuggles combined-system evidence back into the originals.

Answer-Checking Receipt

Did I identify the samples separately before they were combined?
Did I mark the merge event?
Do later labels still preserve separate identity?
What exact object, group or location was measured afterward?
Have I assigned the result to that evidence owner?
Have I avoided splitting a combined value without evidence?
Does my final scientific explanation use only claims the method can support?

Common Traps

Back-assignment trap: giving the combined result to A alone.
Equal-split trap: dividing a combined total equally without evidence.
Identity-erasure trap: forgetting useful pre-combination data.
Uniformity trap: assuming every part of a combined sample has the same state without support.
Label-blindness trap: missing a marker that actually preserves separate tracking.
Average trap: treating a physical combination as if it automatically produced a simple arithmetic average.

Parent and Tutor Teaching Guide

Use two labelled cups or two sets of counters as a representation, without needing a specific science concept. Before combining them, ask the learner to write one fact beside A and one beside B. Then move both sets into one box. Ask: “What facts are still historical facts about A and B? What new observation would belong to the combined box? Could you tell which part came from A if I removed the labels?”

Then repeat while preserving different coloured markers. The learner should see that combination and traceability are separate questions. This is the conceptual repair.

Finally remove the physical aid and test the same reasoning on a table or diagram. Fade support as soon as the learner can preserve evidence ownership independently.

Useful Internal Routes

Authoritative References

Quiet Return

Science evidence always belongs to something. Before a merge, that may be A and B separately. After a merge, it may be the combined system. Track the boundary. Keep what you still know. Refuse to invent what you can no longer separate. That small discipline protects the whole explanation.

Small Group Tutorials