How to Reconcile Two Pieces of PSLE Science Evidence That Seem to Disagree

Wait, what? Two pieces of Science evidence can look as if they disagree and both still be correct.

A learner sees one table saying Set-up A had the higher temperature, then a graph showing Set-up B higher later. Panic follows: “Which one do I trust?” But the real scientific question is often different. Were the two pieces of evidence taken at the same time? Did they measure the same quantity? Were they describing the same object, condition or stage of a process?

Scientific evidence does not become contradictory merely because two numbers, diagrams or statements are different. A conflict exists only when two pieces of evidence make incompatible claims about the same scientific relationship under the same relevant conditions.

Quick Answer

When two pieces of PSLE Science evidence seem to disagree, do not choose one immediately. Put them side by side and check six things: object, condition, time, measured quantity, method and scope.

If one of those differs, the conflict may be only apparent. If all relevant conditions match and the evidence still points in opposite directions, the conflict is real enough to investigate. You may need to inspect measurement quality, repeated results, method limitations, or revise the explanation you were using.

The Exact PSLE Science Learning Job This Guide Owns

This guide teaches how a Primary 5/6 learner should reconcile two pieces of PSLE Science evidence that appear to conflict by deciding whether they really address the same scientific claim.

It is not a general guide to unexpected results, not a graph-reading lesson, and not a method-evaluation article. Those skills support this job, but the dominant task here is evidence reconciliation: how can both pieces be placed into one coherent scientific picture?

Why Learners See Conflicts That Are Not Really There

Science questions often distribute information across different places. A sentence gives the starting condition. A diagram shows a later state. A table gives measurements. A graph shows change over time. One sub-question may refer to the first five minutes while another asks about the final result.

If you flatten all of these into one mental pile, you lose the dimensions that make the evidence meaningful. A result without its time point, condition or measured quantity is incomplete evidence.

This is why the first move is not “Which piece wins?” It is “What exactly does each piece say?”

The Six-Dimension Reconciliation Check

Dimension	Question to ask	Why it matters
Object	Are both pieces about the same object, organism, material or system part?	Different objects can behave differently without contradiction.
Condition	Were the relevant conditions the same?	A changed condition can legitimately change the outcome.
Time	Were both observations made at the same stage or time?	A process can reverse, cross over or accumulate.
Measured quantity	Are both pieces measuring the same thing?	Temperature, mass, height, rate and amount are not interchangeable.
Method	Were the observations or measurements obtained in comparable ways?	Different methods can have different limits or precision.
Scope	Is one statement about one case while the other describes a broader trend?	A local exception does not automatically erase a wider pattern.

If any one of these differs, the two pieces of evidence may be answering different questions.

The Reasoning Law Still Comes First

Use the usual PSLE Science chain:

Read the given information.
Identify the scientific object or relationship.
Separate observation from inference.
Select the relevant concept.
Explain the causal mechanism.
Connect the mechanism to the exact condition.
State the outcome.
Check the outcome against all relevant evidence.

Evidence reconciliation happens at the last step, but it may force you to travel backwards. If a second piece of evidence does not fit, recheck the object, condition, concept or mechanism rather than forcing the evidence to match your first answer.

Worked Example 1: Same Experiment, Different Times

Imagine two containers cooling from different starting temperatures. At five minutes, Container A is still warmer than Container B. At twenty minutes, both have reached the same room temperature.

A learner says, “The data conflict. First A is warmer, then they are the same.”

They do not conflict. The time dimension changed.

Evidence 1 describes an earlier state.
Evidence 2 describes a later state.
A system can change over time, so both observations can fit one coherent process.

The repair is simple: attach the time point to every observation before comparing them.

Worked Example 2: Different Quantities Are Not Opposite Results

Suppose one table shows that Plant A has more leaves than Plant B. Another measurement shows that Plant B is taller.

There is no contradiction. Leaf number and height are different measured quantities. A plant can be taller while having fewer leaves.

The error happens when the learner silently replaces both measurements with a vague idea such as “grew more.” Science becomes clearer when you preserve the measured quantity exactly.

Write the nouns into your scratch work: height, number of leaves, mass, temperature, time. Do not compress them into “better,” “more” or “faster” unless the question actually supports that relationship.

Worked Example 3: A Trend With One Unusual Reading

A student repeats a simple measurement five times. Four readings are close together. One is much higher.

Does the unusual reading “disagree” with the other four? Yes, at the level of the recorded measurement. But that does not tell you why.

You should keep several possibilities alive:

The unusual reading may reflect a measurement or recording problem.
A condition may have changed during that trial.
The system may genuinely vary more than expected.
The four similar readings may themselves be misleading if the method has a shared bias.

At Primary level, you do not need advanced statistics to reason well. You need the discipline to say: the data differ; the cause of that difference must still be investigated.

Worked Example 4: A Diagram and a Sentence That Seem to Disagree

Imagine text stating that a container was sealed at the start of an experiment. A later diagram shows the lid removed while a measurement is taken.

A learner may call this contradictory. But first identify the stage. If the diagram represents the later measurement step, both may be correct: sealed during the experiment, opened afterwards for observation.

The key is chronology. A scientific setup is not always frozen in one state.

Three Kinds of “Conflict”

1. Apparent conflict

The evidence differs because the object, condition, time, quantity or scope differs. No scientific contradiction needs repair.

2. Measurement or method conflict

The pieces should agree, but the method, measurement or recording may have introduced variation. The scientific explanation may still survive, but the evidence quality needs examination.

3. Explanatory conflict

The evidence is genuinely difficult to reconcile with your current mechanism. This is the important case: revise the explanation rather than ignoring the inconvenient evidence.

A Seven-Step Reconciliation Protocol

Write Evidence A literally. Do not interpret yet.
Write Evidence B literally.
Tag each one. Object, condition, time, quantity, method, scope.
Find the shared scientific claim. What relationship would both pieces need to address before they can truly conflict?
Classify the conflict. Apparent, method/measurement, or explanatory.
Repair the model. Add a missing condition, narrow the claim, change the mechanism or keep uncertainty open.
Return to the question. State only what the combined evidence supports.

Failure Signatures: What This Mistake Looks Like on Paper

You cross out a correct earlier answer because a later graph “looks different.”
You use “more” without naming what is more.
You compare measurements taken at different times as if they were simultaneous.
You ignore one piece of evidence because it is inconvenient.
You average together values that measure different things.
You decide one source must be wrong without checking whether the conditions differ.
You force the explanation to stay unchanged even after genuinely contradictory evidence appears.

The Earliest Weak Link

The earliest weak link is usually not “poor Science.” It is often lost context.

Ask the learner to point to each piece of evidence and complete this sentence:

“This evidence shows __________ about __________ under __________ at __________.”

If they cannot fill the blanks accurately, do not yet ask them to reconcile the evidence. Repair evidence reading first.

Do Not Turn Reconciliation Into “Explaining Away” Evidence

There is a dangerous version of this skill: inventing excuses so your favourite answer can never be wrong.

Good reconciliation reduces confusion. Bad reconciliation protects an answer from evidence.

If two genuinely comparable measurements repeatedly disagree with what your explanation predicts, revise the explanation. Do not keep adding imaginary hidden factors unless the question or evidence gives you a reason to consider them.

How to Use This in Open-Ended Questions

When an open-ended question gives more than one source of evidence, build the answer only after reconciliation:

What does each source directly show?
Do the sources refer to the same condition?
Do they measure the same quantity?
Do they describe different stages?
What scientific mechanism can account for the whole evidence set?
Is any evidence left unexplained?

An answer is stronger when every important piece of evidence has a place.

How to Use This During Revision

Take a completed practice question that used a table, graph or several observations. Instead of solving it again, make an evidence map.

Evidence	Object	Condition	Time	Quantity	Job in explanation
A
B
C

This forces you to preserve scientific identity instead of treating “the data” as one vague thing.

Unfamiliar Transfer Test

You understand this skill when you can transfer it across topics. Try four original mini-cases:

A graph and a final measurement appear different because they refer to different times.
Two measurements differ because one is rate and the other is total amount.
One trial differs from repeated trials under apparently similar conditions.
A diagram and sentence appear inconsistent until you realise they show different stages.

For each, say whether the conflict is apparent, methodological or explanatory. Then justify the classification.

Delayed Independent Return

Three to seven days later, take a new multi-source Science question and do this without notes:

Write each evidence statement separately.
Tag object, condition, time and quantity.
Identify any apparent conflict.
Decide whether it is real.
Construct one explanation that respects all the evidence.

If you can do that independently, you are no longer simply reading data. You are coordinating evidence.

Answer-Checking Receipt

I used all relevant evidence, not only the convenient piece.
I preserved what each measurement actually measured.
I checked whether the time or condition changed.
I did not call two different quantities “contradictory.”
If the conflict was real, I revised the explanation rather than hiding the conflict.
My final claim is no stronger than the combined evidence.

Useful Internal Routes

How Do We Know?

The official 2026 PSLE Science syllabus assesses candidates’ ability to interpret and analyse information, evaluate observations, information and methods, and communicate explanations and reasoning. The 2023 Primary Science syllabus also places scientific inquiry across the themes rather than treating evidence use as a separate chapter.

Science education research shows that learners do not always respond productively when information contradicts what they expected. Reviews of conceptual change and contradicting information find that anomalous evidence alone is often insufficient; learners benefit from support that helps them compare claims, consider alternatives and rebuild a more coherent explanation. Research on scientific argumentation likewise treats evidence evaluation and justification as central rather than optional additions to factual knowledge.

This article therefore uses contradiction as a reasoning signal, not a punishment. The goal is not to make children distrust evidence. It is to teach them to preserve the conditions that give each piece of evidence meaning.

Evidence Boundaries

This guide is not an official PSLE marking rubric. It does not claim that every multi-source question contains conflicting evidence, nor that learners must write the six reconciliation dimensions in an examination answer. The protocol is a learning tool for building disciplined interpretation.

Also, “conflicting” evidence does not automatically mean one measurement is wrong. Real systems can change over time, vary across cases and respond differently under changed conditions. Good reasoning keeps those possibilities alive until the evidence separates them.

Teaching Guide for Parents and Tutors

When a learner says, “These two results contradict each other,” resist the urge to explain immediately. Ask six short questions:

Same object?
Same condition?
Same time?
Same measured quantity?
Same method?
Same scope?

Usually one question reveals the earliest weak link. If it is time, practise time-labelled evidence. If it is quantity, practise naming the measured variable. If the conflict remains after all six, then move to method quality or explanation revision.

Do not reward a learner for making every piece of evidence agree. Reward them for noticing when it genuinely does not. Scientific maturity includes the ability to leave a mismatch visible long enough to investigate it.

Authoritative References

Quiet Return

A strong Science learner does not demand that every piece of evidence look the same.

They ask a better question: What does each piece actually say, and can they belong to the same scientific story?

Sometimes the answer is yes because the time, condition or quantity changed. Sometimes the answer is no, and the model must be repaired. Either way, the learner has moved beyond collecting results. They are learning to make evidence live together without distorting it.

Small Group Tutorials