How to Choose the Decisive Evidence for a PSLE Science Answer Without Copying the Whole Table

Wait, What? Copying More Data Can Make an Answer Less Scientific

A table contains twelve numbers. The question asks why Setup A supports a stronger conclusion than Setup B.

A learner copies all twelve numbers into the answer.

The response is long. It looks careful. But the reasoning is buried.

The better answer may need only one comparison: the two values measured under the same relevant condition, or the before–after change that directly answers the question.

Evidence is not the amount of data you copy. Evidence is the information that changes the scientific decision.

This guide teaches how to find that decisive evidence, preserve its scientific meaning, and then connect it to the mechanism rather than treating the number itself as the explanation.

Quick Answer

When a PSLE Science answer needs evidence from a table, graph, diagram or investigation, use this route:

READ THE QUESTION → NAME THE CLAIM → IDENTIFY THE QUANTITY OR OBSERVATION THAT CAN TEST IT → CHOOSE THE MINIMUM FAIR COMPARISON → KEEP UNITS / TIME / REFERENCE → STATE THE EVIDENCE → APPLY THE CONCEPT → EXPLAIN THE MECHANISM → STATE THE OUTCOME → CHECK THAT EXTRA DATA DO NOT CHANGE THE CONCLUSION.

Do not copy every number just because it is present. Do not choose only the largest or smallest number. Do not remove units. Do not compare values taken under different conditions unless the question requires that comparison. And do not stop after quoting the evidence if the question asks for an explanation.

The Exact PSLE Science Learning Job This Guide Owns

This guide owns one learner job: how a Primary 5 or Primary 6 learner selects the smallest sufficient piece or set of evidence from the information supplied by a PSLE Science question, uses the correct comparison reference and units, and connects that evidence to a scientific explanation without copying the whole representation.

It does not replace the guide on extracting all givens before solving. It does not replace the page on combining evidence from several representations. It does not replace trend reading or conclusion strength. Its job is selection:

Which evidence actually decides this answer?

The Current 2026 PSLE Science Frame

For examination from 2026, Standard PSLE Science assesses attainment in the 2023 Primary Science syllabus. SEAB’s assessment objectives include applying scientific facts, concepts and principles, interpreting and analysing information, evaluating observations, information and methods, and communicating explanations and reasoning.

These objectives make evidence selection important. Analysis is not a data dump. It means recognising which information has scientific relevance to the question.

Data Is Not Automatically Evidence

A table may contain:

starting values;
final values;
time points;
several tested conditions;
repeated trials;
different outcomes;
information useful for another sub-question.

The scientific question decides which of those values become evidence for this claim.

The Claim–Evidence Match

Question asks…	Decisive evidence usually needs…
Which setup has the greater final value?	Comparable final values at the same stated condition/time.
Which changed more?	Starting and ending values for each relevant setup.
Which process was faster over the same interval?	Comparable change over equal time.
Does the measured outcome increase with the tested condition?	Several ordered condition–outcome pairs, not one pair.
Which factor caused the difference?	A fair comparison that isolates that factor plus the relevant outcome.
When did the response begin?	Observations immediately before and after onset.
Is the conclusion supported?	The values/observations that directly bear on the conclusion and any exception that could weaken it.

Worked Example 1 — The Whole Table Is Not Needed

Two identical cups begin at 80°C. Their temperatures are recorded:

Time / min	Cup A / °C	Cup B / °C
0	80	80
5	68	72
10	60	66
15	55	62

Question: Which cup has the higher temperature after 15 minutes?

Decisive evidence: Cup A = 55°C; Cup B = 62°C at 15 minutes. Cup B is higher.

You do not need to copy the 5- and 10-minute values.

Now change the question: Which cup cooled more during the first 10 minutes?

Now the decisive evidence changes:

Cup A: 80°C → 60°C, decrease 20°C.
Cup B: 80°C → 66°C, decrease 14°C.

The same table contains different decisive evidence for different claims.

Worked Example 2 — Do Not Select the Biggest Number Automatically

Two wet cloths begin with different amounts of water:

Cloth	Starting water / g	After 20 min / g
P	100	70
Q	60	40

Question: Which cloth lost more water?

Looking only at 70 and 40 is wrong because those are final amounts. The decisive evidence is the change:

P lost 30 g.
Q lost 20 g.

The learner must select evidence for the quantity asked, not the most visually striking number.

Worked Example 3 — One Pair Cannot Prove a Trend

A table shows five light levels and five measured outcomes.

Question: “Describe the relationship between light level and the measured outcome.”

Do not select only the lowest and highest points unless those endpoints truly represent the full pattern. Check the intermediate data. A rise followed by a plateau is not simply “the outcome increases as light increases” across the entire range.

For trend claims, the decisive evidence is a pattern across several ordered conditions, including any plateau, turning point or exception.

Worked Example 4 — Fair Comparison Before Number Quoting

Setup A receives more light and more water than Setup B. A produces more plant growth.

If the question asks whether more light caused the additional growth, quoting the growth values is not enough. The decisive evidence for causation is missing because the comparison changes two conditions.

Evidence selection includes recognising when no available pair can answer the causal question fairly.

Worked Example 5 — Evidence From a Diagram

A circuit diagram shows an open switch and an unlit bulb.

The decisive diagram evidence is not the size of the drawn bulb or the length of the wire. It is the broken conducting path at the open switch, together with the circuit relationship.

Good evidence selection ignores decorative or schematic features that do not carry scientific meaning.

Worked Example 6 — Evidence From an Investigation Method

Two materials are tested as thermal insulators. The cups, starting temperatures, volume of water, time and surroundings are comparable; only the wrapping material differs. Cup X remains warmer.

The decisive evidence is two-part:

the method isolates the wrapping material as the relevant changed factor;
the measured temperature outcome differs under the comparable test.

A number without the fair-comparison structure would not support the same causal claim.

Worked Example 7 — Decisive Evidence Can Be an Exception

Suppose four data points increase, but the fifth drops sharply.

If a learner wants to claim “the outcome always increases as the condition increases”, the fifth point becomes decisive because it contradicts the absolute claim.

Evidence selection is not always about finding the strongest supporting value. Sometimes the scientifically decisive evidence is the one that limits or defeats the claim.

The Minimum-Sufficient-Evidence Rule

Use enough evidence to support the claim, but not so much that the logic disappears.

For many comparison answers:

one aligned pair + the comparison reference + the unit

may be enough.

For a trend, one pair is not enough. For a causal claim, numbers alone may not be enough. The evidence requirement depends on the claim type.

Keep Units Attached to Evidence

“A changed by 20” is incomplete if the quantity matters.

Was it:

20°C?
20 g?
20 mL?
20 seconds?
20 organisms?

The unit is part of the scientific meaning. Keep it when quoting or calculating evidence.

Keep the Comparison Reference Attached Too

“Higher” means nothing without a reference.

Write mentally:

higher than what, at which time, under which condition, for which measured quantity?

Evidence Is Not the Explanation

Weak answer:

“Cup B is a better insulator because it is 62°C and Cup A is 55°C.”

The numbers support the comparison. They do not yet explain the mechanism.

A complete explanation connects the tested material to the rate of thermal energy transfer and then to the temperature outcome, within the conditions of the investigation.

Use:

EVIDENCE → CONCEPT → MECHANISM → CONDITION → OUTCOME.

Do Not Double-Count the Same Evidence

If a graph and a table show exactly the same measurements, quoting both does not create two independent pieces of evidence.

Use whichever representation makes the decisive relationship clearest, unless the question requires both.

When More Than One Evidence Point Is Necessary

To establish a trend across several conditions.
To show a turning point or plateau.
To compare variability across repeated trials.
To show that a result is not a one-off exception.
To combine a method condition with an outcome for a causal claim.
To show that two independent observations support the same mechanism.

When Less Evidence Is Better

If a question asks which setup had the higher value after ten minutes, copying every intermediate time point adds clutter. Select the 10-minute values.

If a question asks which one changed more, calculate the change rather than listing all raw values without interpretation.

Observable Failure Signatures

Failure signature	Earliest weak link	Repair path
Copies the whole table	Claim not identified	State the claim first, then select evidence.
Chooses largest number	Quantity misunderstood	Identify final value, change, rate or count.
Quotes values from different times	Comparison misaligned	Match time and condition.
Leaves out units	Scientific meaning stripped	Carry units with evidence.
Uses one pair to claim a trend	Insufficient pattern evidence	Check several ordered data points.
Uses numbers as mechanism	Evidence/explanation collapsed	Add the scientific causal link.
Ignores an exception	Confirmation-only selection	Check for evidence that limits the claim.

Misconception Repair — “More Numbers Mean More Marks”

Not necessarily. A concise answer that uses the correct comparison can be scientifically stronger than a long list of irrelevant values.

Misconception Repair — “The Decisive Number Is the Biggest One”

The decisive value is the one that answers the claim under the correct reference—not necessarily the maximum or minimum.

Misconception Repair — “Evidence Means Only Numbers”

Evidence can be a labelled diagram feature, a qualitative observation, a fair-test condition, a presence/absence result or a relationship across several measurements.

Misconception Repair — “If I Quoted the Data, I Explained It”

Quoting evidence answers “What supports the claim?” Explanation answers “Why does the scientific relationship produce this outcome?” Keep both jobs distinct.

The Decisive-Evidence Protocol

Underline the command and claim.
Name the measured quantity or observation.
Find the comparison reference.
Align time, condition and object.
Select the smallest sufficient evidence set.
Keep units and direction.
Check for an exception.
Apply the scientific concept.
Explain the mechanism if required.
Remove copied data that do not change the decision.

Practice Sequence

Take one table and write five different questions; select different evidence for each.
Underline irrelevant numbers.
Practise same-time and same-condition comparisons.
Practise final value versus total change.
Practise trend claims that need multiple points.
Practise causal claims where method evidence matters.
Practise one contradictory point that limits a claim.
Write evidence first, then mechanism.
Transfer to diagrams and qualitative observations.

Unfamiliar Transfer Challenge

A mystery device produces this table:

Condition	Reading X	Reading Y
1	4	20
2	6	18
3	9	15
4	9	13

Question A: Which reading is greater at Condition 4? Decisive evidence: X = 9, Y = 13.

Question B: Does X keep increasing across all tested conditions? Decisive evidence includes Conditions 3 and 4: X remains 9, so the claim of continued increase is unsupported.

The device is unknown. The evidence-selection skill survives.

Delayed Independent Return

Three to five days later, choose a new table, graph and diagram. For each, answer:

What is the claim?
What is the minimum evidence?
What comparison reference is required?
What unit matters?
What evidence is irrelevant to this claim?
Does any point weaken the conclusion?
Where does explanation begin after evidence?

The Answer-Checking Receipt

Did I answer the exact claim?
Did I select evidence for the correct quantity?
Did I compare like with like?
Did I keep the same time/condition?
Did I include units?
Did I avoid copying irrelevant values?
Did I check for exceptions?
Did I distinguish evidence from mechanism?
Would removing any quoted value change the argument?

Evidence and Model Limits

There is no universal rule that every PSLE Science answer must quote a certain number of data points. The amount of evidence required depends on the claim and question wording.

This guide therefore teaches a reasoning principle, not a marking template: use the evidence that is scientifically necessary and sufficient for the claim, then explain the mechanism when the question requires explanation.

Useful Internal Routes

Parent and Tutor Teaching Guide

When a child copies an entire table, do not say only “too much”. Ask:

“Which two or three pieces of information would make your conclusion fall apart if they were removed?”

That question forces the learner to identify the load-bearing evidence.

Use one table for several questions. This is especially powerful because the representation stays fixed while the evidence job changes. The learner sees that “read the table” is not one skill; evidence selection depends on the claim.

Then make the child explain after quoting the evidence. If the answer stops at numbers, ask, “What scientific relationship makes those numbers matter?”

Authoritative and Research References

The broader research supports evidence-based reasoning and argumentation. It does not specify an official PSLE rule for how many data values to quote.

The Quiet Ending

The strongest evidence is not the longest evidence.

It is the evidence that connects directly to the question, survives a fair comparison, keeps its scientific meaning, and supports the next step of reasoning.

Find that evidence. Then explain why it matters.

Small Group Tutorials