How to Choose Between Two Plausible Explanations in PSLE Science Using the Evidence

Wait, What? Two Explanations Can Both Sound Scientific

A learner often thinks the difficult part of Science is finding one explanation. A harder and more important problem appears when two explanations both sound possible.

Suppose a wet cloth dries more slowly in one place. Perhaps the air is more humid. Perhaps the air is moving less. Both can affect evaporation. Which explanation should you choose?

The answer is not “whichever fact I remember best”. The answer is: compare what each explanation would predict, then look for the evidence that separates them.

Science becomes stronger when competing explanations are allowed to face the same evidence.

Quick Answer

When two explanations are plausible, use this chain:

FREEZE THE OBSERVATION → STATE EXPLANATION A → STATE WHAT A PREDICTS → STATE EXPLANATION B → STATE WHAT B PREDICTS → FIND THE EVIDENCE THAT DIFFERS BETWEEN THOSE PREDICTIONS → DECIDE WHETHER THE EVIDENCE SUPPORTS A, B, BOTH OR NEITHER → KEEP THE CLAIM WITHIN THE EVIDENCE.

The key idea is discriminating evidence: evidence that would look different if one explanation were true rather than the other.

Owned PSLE Science Learning Job

This guide owns one PSLE Science learner job: how a Primary 5/6 learner compares competing scientific explanations for the same result and uses evidence to decide between them.

Keep the observation separate from the explanation.
Hold two possible explanations in mind without collapsing them too early.
Translate each explanation into a testable prediction.
Identify which evidence would distinguish the explanations.
Recognise when the available evidence cannot decide.
Avoid assuming that one correct fact automatically explains the result.
Use unfamiliar examples without needing a memorised model answer.
Communicate the conclusion at the correct strength.

The Current Official PSLE Science Frame

The 2026 PSLE Science syllabus states that Standard Science assesses attainment in the 2023 Primary Science syllabus. The assessment objectives include application of scientific facts, concepts and principles, interpretation and analysis of information, evaluation of observations, information and methods, and communication of explanations and reasoning.

Comparing explanations belongs naturally inside that frame. The learner is not merely recalling a fact. The learner is asking which explanation best accounts for the evidence under the conditions given.

Why “Both Are Possible” Is Sometimes the Correct Scientific State

Students often feel pressure to choose one answer immediately. But if the evidence does not distinguish two explanations, forcing a choice is poor reasoning.

Imagine two identical-looking seedlings grow poorly. Explanation A: they received too little water. Explanation B: they received too little light. If the question gives no water or light information, both remain possible. The observation “poor growth” does not by itself identify which condition caused it.

Not enough evidence to decide is a scientifically meaningful conclusion when the evidence really is insufficient.

The Eight-Step Competing-Explanation Protocol

Step 1 — Freeze the Observation

Write only what is shown. “The water in Cup A cooled more slowly than the water in Cup B” is an observation about the measured outcome. “Cup A is a better insulator” is already an interpretation.

Step 2 — State Explanation A Clearly

Do not use a vague label such as “heat”. Write a mechanism: “Cup A may lose thermal energy to the surroundings more slowly because its wall material is a poorer conductor of heat.”

Step 3 — Ask What A Predicts

If the material explanation is correct, then changing the material while keeping relevant conditions similar should change the cooling behaviour in a consistent direction.

Step 4 — State Explanation B Clearly

Perhaps Cup A began with a larger volume of water. A larger volume can change how quickly the temperature changes under otherwise similar conditions.

Step 5 — Ask What B Predicts

If volume is the important cause, then two cups made from the same material but containing different volumes should show the relevant difference in cooling behaviour.

Step 6 — Find Discriminating Evidence

Now ask what the question tells you about cup material and water volume. If the volumes are explicitly equal but the materials differ, the volume explanation is weakened or eliminated. If the materials are the same but volumes differ, the material explanation is weakened. If both differ, the experiment cannot cleanly attribute the result to one factor.

Step 7 — Decide Among Four Outcomes

A is better supported.
B is better supported.
Both remain possible.
Neither fits the evidence well.

Step 8 — State the Conclusion at the Right Strength

If one explanation is directly supported and the alternative contradicts a controlled condition, you can write firmly. If both survive, say that the evidence does not distinguish them and identify what extra evidence would help.

Worked Example 1 — Why Did One Cloth Dry Faster?

Two identical wet cloths are hung for one hour. Cloth A dries faster than Cloth B.

Explanation A: More air moved around Cloth A.

Explanation B: Cloth A had a larger exposed surface area.

Both explanations can affect evaporation. To choose, inspect the setup. If both cloths are spread to the same area but one is under a fan, Explanation A is better supported. If the air conditions are the same but Cloth B is folded, Explanation B is better supported. If A is both under a fan and spread out while B is folded in still air, the comparison cannot isolate one cause.

This is the difference between knowing facts and using evidence. The facts tell you what can affect evaporation. The evidence tells you which factor the setup can support.

Worked Example 2 — Why Is a Bulb Dimmer?

Two circuits contain bulbs that appear identical. Bulb X is dimmer than Bulb Y.

Possible explanations could involve the number or condition of cells, the arrangement of components, the bulb itself or the completeness of the conducting path.

Do not select “the battery is weak” simply because that is familiar. Ask what the diagram and description actually show. If both circuits use identical new cells but one arrangement places more bulbs in a way that changes the energy delivered to each bulb, then the arrangement explanation has direct support. If cell condition is not shown, it may remain possible but unverified.

Worked Example 3 — Why Did a Plant Lose More Water?

A plant in Setup P loses more water over the same period than a plant in Setup Q.

Explanation A: P has more leaves, giving a larger total leaf surface through which water can be lost.

Explanation B: P is placed in moving air, which increases the rate of water loss from leaves.

If leaf number and leaf size are controlled but only air movement differs, B is the stronger explanation. If air movement is controlled but leaf area differs, A is stronger. If both differ, both may contribute and the experiment cannot attribute the difference uniquely.

Worked Example 4 — A Surprising Experimental Result

Three repeated measurements are close together, but a fourth is much larger.

Explanation A: The fourth measurement contains a reading or timing error.

Explanation B: A real condition changed during the fourth trial.

The unusual value alone does not identify which explanation is correct. Useful discriminating evidence could include checking the instrument, inspecting the recorded method, repeating the trial under controlled conditions, or identifying whether a relevant environmental condition changed.

The correct scientific response is not to delete the result automatically. It is to investigate why it differs.

What Counts as Discriminating Evidence?

Evidence is discriminating when it produces different expectations under the competing explanations.

If A predicts a larger value and B predicts no difference, a clear larger value supports A more strongly.
If A predicts an effect only when light is present and B predicts the effect in darkness too, a dark-condition result can separate them.
If A requires a closed circuit and B does not, checking circuit continuity can distinguish the explanations.
If both explanations predict the same observation, that observation cannot choose between them.

The “Same Prediction” Trap

Suppose Explanation A and Explanation B both predict that the plant will become shorter after severe damage. Observing shorter height does not distinguish A from B. You need a second observation on which their predictions differ.

This is a powerful idea: evidence can support an outcome without identifying the mechanism.

The “One True Fact” Trap

A fact can be correct but still fail to explain the result.

“Plants need water” is true. It does not explain a difference between two plants unless water conditions are relevant to the comparison.

“Metals conduct heat” is true. It does not explain why one metal object warms faster than another unless the arrangement and material properties support that inference.

The “First Explanation Wins” Trap

The first explanation that comes to mind has a psychological advantage because it arrives first, not because it has more evidence.

Repair this by forcing a second plausible explanation internally before answering. You do not always need to write the second one. Its purpose is to test whether your first explanation really fits the evidence uniquely.

Explanation Comparison Table

Question	Explanation A	Explanation B
What mechanism is proposed?	State it clearly	State it clearly
What should we observe if it is correct?	Prediction A	Prediction B
Which given condition matters?	Relevant condition	Relevant condition
What evidence would separate them?	Find a result predicted differently by A and B
What does current evidence justify?	A / B / both / neither

Earliest Weak-Link Diagnosis

You cannot think of a second explanation: concept flexibility may be weak.
You produce two explanations but cannot compare them: prediction reasoning is weak.
You choose from familiarity: evidence selection is weak.
You ignore controlled conditions: inquiry reasoning is weak.
You say “both” even when one contradicts the data: elimination is weak.
You force one answer when evidence cannot decide: claim calibration is weak.

Misconception Repair: “More Evidence” Is Not Always “More Useful Evidence”

Ten observations that both explanations predict do not necessarily separate them. One carefully chosen observation on which they predict different outcomes may be far more useful.

This is why good experiments do not only collect more data. They collect data that can answer a specific scientific question.

Inquiry Protocol: Design the Next Test

If the current evidence cannot choose between A and B, design the smallest fair test that separates them.

What variable must differ between the explanations?
Which condition should be changed?
Which conditions should be kept similar?
What should be measured?
What result would favour A?
What result would favour B?

Do not design a complicated experiment if one simple comparison can distinguish the mechanisms.

How This Helps Multiple-Choice Reasoning

In MCQ, two options may both contain true Science. The correct choice is the one that best fits the exact evidence and condition.

Ask what each option predicts. Then test those predictions against the diagram, table or setup. This is stronger than asking which option “sounds more scientific”.

How This Helps Open-Ended Answers

Open-ended questions may reward one well-supported explanation rather than a list of possibilities. Internally comparing alternatives helps you choose the explanation that is most tightly anchored to the question.

If the question asks for one possible explanation and the evidence cannot decide uniquely, say so by using calibrated language rather than pretending certainty.

Unfamiliar Transfer Example — A Mystery Box

A sealed box makes a clicking sound when tilted. You cannot see inside.

Explanation A: a loose solid object rolls and strikes the wall. Explanation B: a hinged part swings and strikes the wall.

The sound alone is compatible with both. To distinguish them, you might compare the timing, direction dependence, repeated pattern or response to slow versus fast tilting. The exact test must be safe and appropriate, but the reasoning principle is the same: find evidence that the explanations predict differently.

Retrieval and Practice Sequence

Take one result and write two explanations.
For each explanation, write one prediction.
Circle where the predictions differ.
Identify the evidence needed to separate them.
Return later with a changed context and repeat.

Delayed Independent Return Test

Two days later, use an unfamiliar setup. Without hints, can you keep two explanations alive, state the prediction of each, identify the discriminating evidence and refuse to overclaim if the evidence does not decide?

If yes, you are moving from answer recall toward scientific judgement.

Answer-Checking Receipt

Did I separate the observation from the explanation?
Are both explanations actually scientifically plausible?
What does each explanation predict?
Which given evidence distinguishes them?
Did I accidentally use a fact that is true but irrelevant?
Does one explanation contradict a stated condition?
If evidence cannot decide, did I say so?
Did I state the outcome at the correct strength?

Parent and Tutor Teaching Guide

When a child gives an explanation, ask: “What else could cause the same result?” Then ask: “What would we observe differently if your first explanation were correct?”

This is usually more powerful than asking for a second answer alone. The goal is not brainstorming. The goal is comparison under evidence.

If the child cannot generate a competing explanation, offer one that is scientifically plausible but not obviously correct, then ask the child to decide what evidence would separate the two. This trains evaluation rather than dependence.

Useful Internal Routes

Authoritative External References

The Quiet Ending

A strong Science learner does not fall in love with the first explanation. They ask what else could produce the same result, what each explanation predicts, and which observation would separate them.

That is a deeper form of confidence. You are no longer hoping the familiar answer is right. You are giving the explanations the same evidence and watching which one survives.

Small Group Tutorials