Wait, What? If A is greater than B and B is greater than C, you may be able to rank A, B and C—but only if every comparison refers to the same scientific quantity under a compatible basis.
PSLE Science questions can distribute evidence across several sentences, diagrams, tables or sub-questions. One comparison may tell you that Set-up P has a larger measured outcome than Q. Another may compare Q with R. A learner may then rush to arrange all three from greatest to least.
Sometimes that is scientifically justified. Sometimes the comparisons use different times, quantities, conditions or reference points, so the chain cannot be joined. And sometimes the evidence supports only part of a ranking. The skill is not “sort the letters.” The skill is build only the ordering that the evidence actually supports.
Quick Answer
Before joining two pairwise comparisons, check that they refer to the same measured quantity, unit or meaning, relevant time or stage, and scientific basis. If they do, you may connect them. If a link is missing, keep the ranking partial rather than inventing the missing relation.
PAIRWISE EVIDENCE P > Q Q > R Before concluding P > Q > R, check: SAME quantity? SAME comparison basis? COMPATIBLE time/stage? COMPATIBLE conditions? NO hidden change of reference?
The Exact PSLE Science Learning Job This Guide Owns
This guide teaches Primary 5/6 learners to combine multiple pairwise scientific comparisons into a defensible ordering. It is not a generic Mathematics ranking page. The dominant job is PSLE Science evidence reasoning: preserve what is being compared, decide whether separate pieces of evidence can be connected, and stop when the evidence no longer determines the order.
Why Pairwise Evidence Is Easy to Misread
Our eyes like complete lists. If a question gives several objects, we expect there to be a neat first, second, third and fourth. Science does not promise that. Evidence may establish only some relationships.
Three errors are common:
- Direction reversal: reading P > Q as Q > P.
- Basis drift: joining a comparison of temperature with a comparison of time taken, or a final value with a change.
- Gap filling: inventing a relation between two set-ups that were never validly compared.
The Scientific Address of a Comparison
Every comparison needs an address. Ask:
| Part | Question |
|---|---|
| Object | Which set-ups, specimens or cases are compared? |
| Quantity | What exactly is greater, smaller, faster, longer or higher? |
| Time | At the same elapsed time, at the end, or over a stated interval? |
| Condition | Under what tested condition does the comparison hold? |
| Reference | Greater than what? |
| Evidence | Which observation, value or relationship supports it? |
Only comparisons with compatible addresses should be chained.
Worked Example 1: A Valid Three-Set-Up Ranking
Consider an original practice table. The same quantity is measured after the same amount of time for three set-ups.
| Comparison | Evidence |
|---|---|
| P vs Q | P has a larger measured value than Q |
| Q vs R | Q has a larger measured value than R |
If the quantity, units, time and conditions are genuinely the same, you can conclude:
P > Q Q > R therefore P > Q > R
The important word is therefore. It is justified only because Q is the same scientific middle case in both comparisons and the comparison basis has not changed.
Worked Example 2: A Chain That Must Not Be Joined
Suppose P has a greater final temperature than Q, while Q takes longer than R to reach a target temperature. The learner writes P > Q > R.
That ranking is meaningless because the first comparison concerns final temperature while the second concerns time to reach a target. The symbols look similar, but the scientific quantities are different.
Repair: write the quantity beside every comparison before chaining it.
P > Q [final temperature] Q > R [time taken] DO NOT JOIN
Worked Example 3: Partial Ranking Is a Valid Scientific Answer
Suppose the evidence establishes P > Q and R > Q, but gives no valid comparison between P and R.
You know both P and R are greater than Q. You do not know whether P is greater than R, R is greater than P, or they are equal.
P > Q R > Q Known: P and R are both above Q. Unknown: P ? R
This is not an incomplete learner. It is a correct evidence limit.
Ties Matter Too
If the evidence shows P = Q and Q > R under the same basis, then P = Q > R. Do not force P and Q into separate positions merely because a ranking task looks as though every item should have a unique place.
However, equal displayed values do not always prove identical underlying behaviour. Check measurement resolution, rounding and whether the equality refers to one time point or the entire process.
A Ranking Is Not Yet a Causal Explanation
Ordering the results tells you what the evidence shows. It does not automatically tell you why. If the question asks for an explanation, continue the PSLE Science reasoning chain:
READ / OBSERVE → IDENTIFY THE QUANTITY → BUILD THE VALID COMPARISON → SELECT THE RELEVANT CONCEPT → EXPLAIN THE CAUSAL MECHANISM → CONNECT TO THE CONDITION → STATE THE OUTCOME → CHECK AGAINST THE EVIDENCE
A ranking is evidence structure. The mechanism is the scientific explanation.
When Transitive Reasoning Is Safe—and When It Is Not
Transitive reasoning is safe when the comparison has the same meaning throughout. If P has a greater value than Q and Q has a greater value than R for the same scalar quantity under aligned conditions, P has a greater value than R.
It becomes unsafe when:
- the quantity changes;
- the units or denominator change;
- the comparison time changes;
- one statement concerns total while another concerns per-object value;
- one relationship is causal and the next merely correlational;
- the scientific object represented by the same letter changes between parts;
- the conditions differ in a way that affects the outcome.
MCQ Use: Do Not Rank More Than You Need
In a multiple-choice item, an option may present a complete ranking even though the evidence fixes only two relations. Test the uncertain pair. If the question never provides a valid link, reject any option that pretends the order is known.
You also do not need to construct a full ranking if one pairwise comparison is enough to eliminate an option. Scientific efficiency means using the smallest valid evidence set that answers the job.
Open-Ended Use: State What Is Known and Unknown
If evidence is incomplete, a precise answer can say that two set-ups are both greater than a third while their order relative to each other cannot be determined from the information given. That is a stronger scientific habit than guessing a neat sequence.
Failure Signatures
- You write a full greatest-to-least list even though one pair was never compared.
- You join comparisons that refer to different quantities.
- You forget that one value is a total and another is per specimen.
- You carry a ranking from one time point into a later time after the lines cross.
- You reverse one inequality while copying it.
- You assume equal values mean identical systems.
- You use a ranking as if it were already a causal explanation.
Earliest Weak-Link Diagnosis
| Wrong pattern | Earliest weak link | Repair |
|---|---|---|
| Complete order invented | Evidence limit | Mark every unresolved pair with ?. |
| Different quantities chained | Quantity identity | Write the measured quantity beside each comparison. |
| Direction reversed | Comparison tracking | Translate symbols back into words before combining. |
| Old order carried to a new time | Time scope | Rebuild the ranking for the requested time. |
| Ranking used as explanation | Evidence vs mechanism | Add the relevant concept and causal link. |
The LINK Protocol
This is a learning aid, not an official exam method.
- L — Label the scientific quantity and time for each comparison.
- I — Identify the common middle case.
- N — Notice changes of unit, condition, reference or scope.
- K — Keep unknown pairs unknown unless evidence closes the gap.
Original Practice 1: Four Set-Ups, One Missing Link
You are told, for the same measured quantity and time: A > B, B > C, and D > C. What can you rank?
- A > B > C is supported.
- D > C is supported.
- A versus D is unknown.
- B versus D is unknown.
A complete four-item order is not justified.
Original Practice 2: Same Letters, Different Time
At 5 minutes, P > Q. At 20 minutes, a graph shows Q > P. If a later question asks for the order at 20 minutes, the earlier ranking has expired. Re-read the requested time before using any stored comparison.
Original Practice 3: Evidence Network
Draw circles for W, X, Y and Z. Add an arrow from the greater value to the smaller value whenever the question gives a valid pairwise comparison. Do not add an arrow for a pair you merely expect. Then see which orderings can be reached through valid chains.
This visual network is especially useful when information is spread across multiple sub-questions.
Retrieval and Practice Sequence
- Practise three-set-up rankings with one quantity and one time.
- Add ties.
- Add one missing pair so the correct result is only a partial order.
- Add a second quantity and practise refusing to join incompatible comparisons.
- Add a time change or crossing trend.
- Return after several days and solve a mixed set without the LINK prompt.
Unfamiliar Transfer Test
Create four fictional set-ups and five pairwise statements. Make exactly one statement incompatible by changing its quantity or time. Give the set to another learner and ask them to identify which links can form one ranking and which link must remain separate. Then swap roles.
Delayed Independent Return Test
Three to seven days later, use a fresh diagram or table and build the comparison network from scratch. Your receipt is not speed alone. It is whether every edge in the network can be defended from a specific piece of evidence.
Answer and Checking Receipts
- I can name the quantity behind every greater-than or less-than claim.
- I can keep direction correct.
- I can identify the shared middle case.
- I can refuse to join comparisons with different bases.
- I can preserve ties.
- I can mark an unresolved pair as unknown.
- I can distinguish a ranking from the scientific explanation for that ranking.
Parent and Tutor Teaching Guide
Use three cards labelled A, B and C. Give the learner only two comparisons and ask them to place the cards. Then deliberately remove one comparison and ask whether a complete order is still possible. The goal is to make “unknown” feel like a legitimate scientific state rather than a failure to finish.
Next, change the quantity on one card from final value to time taken. Ask whether the chain still works. Finally, ask for a causal explanation after the ranking is correct so evidence structure and scientific mechanism remain separate jobs.
Useful Internal Routes
- Track a changing comparison reference
- Compare cases through a common reference
- Keep time-to-reach separate from value-after-time
- PSLE Science Learning Guide hub
- PSLE Primary Science archive
Official Frame and Evidence Limits
The 2026 PSLE Science syllabus states that the examination assesses knowledge with understanding and application of knowledge and scientific inquiry, including interpreting and analysing information, evaluating observations and information, and communicating explanations and reasoning. The MOE 2023 Primary Science syllabus also frames data as evidence to be analysed for patterns and relationships. This guide teaches that evidence job without inventing an official ranking rule or marking phrase.
- SEAB: 2026 PSLE Science syllabus
- SEAB: PSLE formats examined in 2026
- MOE: 2023 Primary Science teaching and learning syllabus
Quiet Return
A strong Science learner does not force every set of evidence into a neat complete order. They preserve the quantity, time, condition and reference; connect only compatible comparisons; and leave genuine unknowns open. That is not hesitation. It is scientific control.
