How to Track a PSLE Science Sample After It Is Split Into Two Test Portions

Wait, What? If one sample becomes two portions, the two portions are both connected to the same past—but they do not share every future result.

That distinction is easy to lose in a long PSLE Science investigation. A question may describe one original sample, divide it, treat Portion A in one way and Portion B in another way, then show observations or measurements. If you mentally reset the story at the split, you may forget useful starting evidence. If you do the opposite and keep treating both portions as one object, you may copy a result from A into B. Good scientific reasoning needs both truths at once: shared history before the split; separate evidence after the split.

Quick Answer

Draw the investigation as a fork. Write the facts that apply before the fork above it. Those facts may legitimately travel into both branches if the question gives no reason for them to change. After the fork, keep every condition, observation and measurement on its own branch unless the question explicitly combines the portions again.

ORIGINAL SAMPLE
shared starting state / shared earlier evidence
                 |
              SPLIT
              /   \
        PORTION A  PORTION B
        condition A condition B
        result A    result B

The core checking question is: “Did this information exist before the split, or was it produced after the split?”

The Exact PSLE Science Learning Job This Guide Owns

This guide teaches one learner job: how to preserve evidence provenance when one original scientific sample, group, object or batch is divided into two test portions. It does not teach the underlying Biology, Chemistry or Physics concept as a new canonical owner. The scientific topic may change; the reasoning structure remains the same.

You are learning to answer four questions accurately:

What was already true or measured before the sample was divided?
Which later condition applies only to Portion A or only to Portion B?
Which later result belongs to which portion?
What can be compared fairly after the split?

Why This Is Hard: Identity Changes Without a Complete Reset

A split creates a special reasoning situation. Before the split there is one tracked object. After the split there are two tracked portions. The portions inherit a common history, but from the split onward they can experience different conditions. This is not the same as two unrelated samples that began separately, and it is not the same as two samples being combined into one later system.

The useful mental model is inherit, then diverge.

Inherit: each new portion begins with the scientifically relevant history of the original sample, unless the split itself changes a property the question cares about.
Diverge: once different treatments begin, later evidence must stay attached to the branch that produced it.

This resembles keeping a scientific ledger. You do not erase what happened before the fork. You also do not allow later entries in one branch to leak into the other.

The Reasoning Chain

For this kind of question, train the full PSLE Science reasoning chain:

READ GIVEN INFORMATION
→ IDENTIFY THE ORIGINAL SAMPLE
→ MARK THE SPLIT POINT
→ SEPARATE PRE-SPLIT FROM POST-SPLIT INFORMATION
→ ATTACH EACH NEW CONDITION TO A OR B
→ ATTACH EACH RESULT TO THE PORTION THAT PRODUCED IT
→ SELECT THE RELEVANT SCIENCE CONCEPT
→ EXPLAIN THE MECHANISM UNDER THAT PORTION'S CONDITION
→ STATE THE OUTCOME
→ CHECK AGAINST THE EVIDENCE AND THE OTHER BRANCH

Worked Example 1: One Starting Sample, Two Later Conditions

Imagine an original, made-up investigation. A student prepares one batch of identical paper strips and measures a property of the batch under the same starting condition. The batch is then divided into Portion A and Portion B. A is placed under Condition X. B is placed under Condition Y. After the same duration, the student records a different outcome for each portion.

Before thinking about the science concept, organise the evidence:

Information	Scope	Reason
Original preparation method	Both portions	It happened before the split.
Starting measurement	Both portions only if the measurement genuinely describes the original material relevant to both	It was obtained before the split.
Condition X	Portion A	Applied after the split to A.
Condition Y	Portion B	Applied after the split to B.
Final result A	Portion A	Observed from A after its treatment.
Final result B	Portion B	Observed from B after its treatment.

A common mistake would be to say, “Both portions had Final Result A because they came from the same sample.” That confuses shared origin with shared outcome. Another mistake would be, “Nothing before the split matters because A and B are now separate.” That throws away valid starting evidence.

Worked Example 2: The Split Itself Can Matter

Suppose one 100 g batch is divided into two portions, but the question states that Portion A has 40 g and Portion B has 60 g. The original batch’s identity is shared history, but the amount is no longer the same in each branch. A learner must not blindly copy “100 g” onto both portions.

This reveals an important limit: pre-split information travels forward only when its scientific meaning survives the split. A material type may survive. A preparation method may survive. A total amount usually does not survive unchanged when the object is divided. If a property depends on size, amount, surface area or position, you must reread what the split changed.

Therefore the rule is not “copy everything before the fork to both branches.” The rule is “carry forward only the scientifically valid inherited information.”

Observation Is Not Inference

If Portion A changes colour and Portion B does not, those are branch-specific observations. You may then use relevant scientific knowledge to infer or explain what the observations mean. Do not quietly turn an inference about A into a fact about B merely because the two portions shared an origin.

Keep this distinction visible:

Observation: what was directly seen or measured for that portion.
Inference: what you conclude from the observation plus scientific knowledge.
Condition: what the portion experienced.
History: relevant information inherited from before the split.

Failure Signatures: What a Split-Sample Error Looks Like

You may have this weakness if you repeatedly do one of the following:

apply Condition A to both portions;
copy Result A onto Portion B;
forget a starting measurement that genuinely applies to both;
treat the two post-split portions as if they were independent before the split;
carry the original total quantity into both branches after division;
compare A and B at different times without noticing;
combine evidence from A and B into one conclusion when the question asks about only one portion;
say the split proves the two portions are identical in every property.

Find the Earliest Weak Link

Do not repair the final sentence first. Find where the tracking broke.

Wrong answer pattern	Likely earliest weak link	Repair
Uses B’s condition to explain A	Condition scope	Label every post-split condition A or B before reasoning.
Forgets common starting state	Temporal provenance	Write PRE-SPLIT above the fork and preserve valid inherited facts.
Copies a result across branches	Evidence ownership	Write each result under the portion actually observed.
Copies original total into both portions	Quantity meaning	Ask whether the split changed the quantity itself.
Explains with a correct concept but wrong branch	Object identity	Name the portion in every causal step until tracking is stable.

A Fast Question-Reading Protocol: FORK

Use this as a learning protocol, not as an official examination rule.

F — Find the original object. What existed before division?
O — Observe the fork. At what exact step does one become two?
R — Route conditions and results. Which branch owns each later fact?
K — Keep only valid inheritance. Which earlier facts remain true for both portions after division?

Then perform the ordinary scientific chain: evidence → concept → mechanism → condition → outcome → evidence check.

Original Practice: Build the Evidence Ledger

A jar contains a well-mixed collection of small objects. Before testing, a learner records the common preparation method. The collection is divided into equal portions P and Q. P is placed under one stated condition; Q under another. After ten minutes, P is measured twice and Q is measured twice.

Without solving any science topic, make a ledger with five rows:

shared pre-split preparation;
what the split itself changes;
P-only condition;
Q-only condition;
P and Q results kept separate.

Now ask: if the question later gives a new observation only for P, may you add it to Q? No—unless the question gives an independent scientific reason that the observation also applies to Q. Shared ancestry is not enough.

What If the Two Portions Are Supposed to Be Similar?

Similarity helps the fairness of a comparison, but similarity is not a licence to invent identical results. If a mixed original batch is divided carefully, the purpose may be to make the portions comparable at the start. The final evidence must still be read separately.

Also notice the language limit: a question may say two portions are equal in amount, but that does not automatically state that they are identical in every hidden feature. Use only the properties the question establishes or the science justifies.

When Tables and Diagrams Hide the Fork

A split may appear as a sentence, a branching diagram, two table columns or two labelled containers. Do not rely on visual style. Reconstruct the timeline:

same source → split → different condition(s) → later observation(s)

If the table lists P and Q in separate columns, ask whether those labels began before or after division. If a diagram shows arrows from one container to two, ask what is conserved, what is divided, and what later becomes branch-specific.

Misconception Repair: “Same Source Means Same Result”

The misconception sounds reasonable because common origin often helps establish a fair starting point. The missing distinction is time. Common origin supports similarity before different treatments. Once branch-specific conditions act, different outcomes can be exactly what the investigation is designed to detect.

Repair the model with one sentence: “Same before the fork does not mean same after different conditions.” Then test it on a different scientific context so the repair is not tied to one example.

Retrieval and Practice Sequence

Day 1 — Map: take three made-up split-sample diagrams and mark shared history versus branch-specific information.
Day 1 — Explain: say aloud why one pre-split fact may travel forward while one post-split result may not cross branches.
Day 2 — Remove labels: redraw a problem without colours or familiar object names and track only Source → A/B → conditions → results.
Day 4 — Change representation: convert a branching diagram into a table, then reconstruct the fork from the table.
Day 7 — Mixed return: solve a question containing both a split and a later comparison without being told that split-tracking is the target.

Unfamiliar Transfer Test

Try this without looking back. One original set of objects is measured. It is divided into R and S. R receives Treatment 1. S receives Treatment 2. Later, only R is moved to another location and measured again. Which information can be safely used to describe S?

Your answer should separate: (a) valid shared pre-split facts, (b) S’s own post-split condition and evidence, and (c) R-only later information that must not be transferred to S. If you can do that in a new context, the reasoning is becoming transferable.

Delayed Independent Return Test

Three to seven days later, close this guide. On a blank page, draw one original sample splitting into two. Write one fact that can validly be inherited by both branches, one quantity that may change because of the split, and one later result that must remain branch-specific. Then explain the rule without using the words “because the guide says so.”

A strong receipt is not remembering the FORK letters. A strong receipt is correctly tracking a new scientific situation.

Answer and Checking Receipts

I can point to the exact split point.
I can name what information existed before the split.
I can explain why each post-split condition belongs to A, B or both.
I can keep later observations attached to the correct branch.
I can detect when a pre-split quantity no longer survives unchanged after division.
I can compare the branches using the same quantity and time.
I can state what the evidence does not establish.

Common Traps

Trap: treating the split as a complete reset. Repair: preserve valid shared history.
Trap: treating shared history as shared future. Repair: route post-split evidence by branch.
Trap: assuming “equal portions” means identical in every property. Repair: use only what equal refers to.
Trap: transferring an inference across branches. Repair: return to the direct observations for each portion.
Trap: copying an original total into both portions. Repair: preserve quantity meaning after division.

Parent and Tutor Teaching Guide

Do not begin by giving the child the science answer. Give a pencil and ask the learner to draw the fork. Then ask four short questions: “What existed before the split?” “What changed at the split?” “What happened only to A?” “What happened only to B?”

If the learner knows the topic but still swaps results, the weak link is probably tracking rather than concept knowledge. Keep the science simple while repairing the tracking job. Once the child can protect evidence ownership, return to a normal PSLE Science question where the scientific mechanism matters.

A useful teaching variation is to make the two portions visually identical. This prevents colour or position from doing the tracking for the learner. Later, change labels or representations. The scientific identity should survive the cosmetic change.

Useful Internal Routes

Official Frame and Evidence Notes

The current Singapore Primary Science syllabus treats scientific inquiry as involving the interpretation and analysis of information, evaluation of observations and methods, and communication of explanations and reasoning. The 2026 PSLE Standard Science syllabus assesses the 2023 Primary Science syllabus. This guide therefore teaches evidence tracking and reasoning; it does not invent a marking phrase or claim that PSLE candidates must draw a fork in the examination.

Quiet Return

A split-sample question becomes manageable when you stop seeing two containers and start seeing one scientific history that branches. Protect what is genuinely shared. Protect what becomes separate. Then let the evidence—not the drawing position, label colour or familiarity—tell you what belongs where.

Small Group Tutorials