How to Perform in PSLE | Learner’s Guide Vol 0094 | Mathematics: Hold One Quantity Fixed Before Comparing Rows in a Table

Chapter index

1. Define the performance job

The exact job in this volume is to compare two rows only after identifying which quantity changed, which quantity stayed fixed and which result is being judged. That may sound like a small organisational step, but it prevents a large class of errors. Many weak answers contain relevant facts and correct calculations. What they lack is a controlled relationship between those facts.

Treat every comparison as a sentence with hidden blanks: “When ____ changed while ____ stayed fixed, ____ happened.” If the learner cannot fill the blanks, the answer may be comparing unlike cases or assigning a role to the wrong detail.

This does not mean every question is an experiment. English paragraphs, Mathematics tables and Science investigations all organise relationships differently. The shared habit is to locate the reference that makes a change meaningful before explaining the result.

A comparison is only as clear as its baseline. Name what changed, what stayed fixed and what the result actually shows.

Watch especially for rows with several numerical columns, repeated labels, totals, rates, averages or different group sizes. These words invite comparison, but they do not identify a fair basis by themselves. The learner must recover the baseline, denominator, time interval, argument direction or controlled condition.

The aim is not longer answers. It is a shorter reasoning chain that another reader can reconstruct. When that chain is stable, the learner can write with more confidence and spend less time repairing entire solutions.

Back to contents · Next: see the mechanism

2. See the mechanism

Use a C–F–R audit: changed quantity, fixed quantities and recorded result. A fair one-factor comparison changes the quantity of interest while holding the relevant conditions constant. When that is impossible, state the limitation or choose a different pair of rows. This structure turns a vague comparison into a sequence. It also helps locate failure. If the result is wrong, inspect whether the changed quantity was misidentified, the fixed reference was not actually fixed, or the conclusion exceeded the recorded outcome.

Imagine comparing photographs taken with different zoom levels. An object may look larger because it changed, because the camera moved closer or because the image was cropped. A trustworthy comparison controls the viewpoint. In examination work, the “viewpoint” may be a denominator, a starting value, a time interval, a paragraph claim or an experimental condition.

The fixed element does not have to be numerically equal in every task. Sometimes the common basis is created by conversion: cost per item, percentage of each class or change per minute. The learner must state the conversion and use it consistently.

Likewise, not every observed difference is causal. Science requires a fair design; Mathematics requires comparable quantities; English requires a link between language choice and paragraph direction. The mechanism disciplines the conclusion without forcing all subjects into identical wording.

A useful backwards check starts from the final sentence. Underline the result. Ask what comparison produced it. Then point to the changed feature and the fixed reference. If any step cannot be located, the conclusion needs repair or qualification.

This audit is especially valuable after a learner has acquired topic knowledge. It converts “careless” into a named routing error. A named error can become a visible trigger for future practice.

Previous chapter · Contents · Next: six-step launch

3. Launch with six steps

Use the following launch during guided practice. Write the labels at first. After several successful returns, shorten them to margin marks or mental questions.

• Step 1. Name the comparison question before touching the numbers.
• Step 2. Mark the column that is supposed to change.
• Step 3. Check every other relevant column for equality or a justified conversion.
• Step 4. Choose rows that isolate the intended comparison.
• Step 5. Calculate the requested difference, ratio or rate.
• Step 6. Write the conclusion with the condition that was held fixed.

The order matters. Learners often calculate or explain before deciding what comparison the question asks for. By naming the job first, they reduce the chance of performing correct work on the wrong pair of values or the wrong sentence.

During oral rehearsal, require one complete sentence: “I am comparing ___ with ___ while keeping ___ fixed.” This sentence may not appear in the final answer, but it exposes whether the learner understands the structure.

Next ask for a counter-comparison. Which tempting row, sentence or setup should not be paired, and why? Rejecting a bad comparison builds stronger control than copying a valid one because the learner must name the disqualifying difference.

Finally, compress. The final answer should state the requested result and the decisive basis. Do not paste the entire audit into a one-mark response. The audit is a thinking route whose output must fit the task.

Move to speed only when the learner can explain the route accurately. Timed repetition of a wrong route produces faster wrong answers.

Previous chapter · Contents · Next: worked repair set A

4. Worked repair set A

The first set is guided. Predict the failure before reading the diagnosis. The goal is to see why the tempting move feels reasonable and exactly where it becomes illegal.

1. Same time, different distance

Situation. A table gives distances covered by two cyclists after 10 minutes and 15 minutes.

Tempting move. Compare the 10-minute distance of one cyclist with the 15-minute distance of the other. This is attractive because it uses a visible detail quickly, but speed has replaced comparison design.

Diagnosis. Both cyclist and time change, so the distance difference cannot isolate performance. Name the exact extra change, missing reference or role error. Avoid the blanket label “careless”; it cannot tell the learner what to notice next time.

Repair. Compare both cyclists at 10 minutes, or calculate a justified rate if the relationship and task permit it. The repair keeps useful work and changes only the unsupported comparison or conclusion.

Check. Write the fixed condition—same time interval—beside the selected rows. Read the answer backwards: result, comparison, changed feature and fixed reference. Each part should be recoverable from the task.

Transfer. Add a 12-minute row and decide whether interpolation is authorised. Solve the changed case without the model beside you, then state why the altered detail changes—or preserves—the legal comparison.

• Mark the changed feature.
• Name the fixed reference.
• State the recorded result.
• Reject one unfair comparison.
• Create one changed-detail retest.

2. Same price, different quantity

Situation. Packets of rice have different masses and total prices.

Tempting move. Choose the cheapest total price. This is attractive because it uses a visible detail quickly, but speed has replaced comparison design.

Diagnosis. A smaller packet may cost less overall while costing more per kilogram. Name the exact extra change, missing reference or role error. Avoid the blanket label “careless”; it cannot tell the learner what to notice next time.

Repair. Convert each option to the same unit quantity before comparing value. The repair keeps useful work and changes only the unsupported comparison or conclusion.

Check. The fixed basis is one kilogram, not one packet. Read the answer backwards: result, comparison, changed feature and fixed reference. Each part should be recoverable from the task.

Transfer. Change one packet to a promotional bundle and include every paid and free unit. Solve the changed case without the model beside you, then state why the altered detail changes—or preserves—the legal comparison.

• Mark the changed feature.
• Name the fixed reference.
• State the recorded result.
• Reject one unfair comparison.
• Create one changed-detail retest.

3. Same total, different parts

Situation. Two classes each have 40 pupils but different numbers choosing chess.

Tempting move. Compare the fractions using only the numerators. This is attractive because it uses a visible detail quickly, but speed has replaced comparison design.

Diagnosis. Because totals are equal, the numerators can be compared directly here, but the reason must be known. Name the exact extra change, missing reference or role error. Avoid the blanket label “careless”; it cannot tell the learner what to notice next time.

Repair. With the same class size, the class with more chess pupils also has the larger fraction. The repair keeps useful work and changes only the unsupported comparison or conclusion.

Check. State why equal denominators make the direct comparison legal. Read the answer backwards: result, comparison, changed feature and fixed reference. Each part should be recoverable from the task.

Transfer. Change one class total to 50 and rebuild the comparison with fractions or percentages. Solve the changed case without the model beside you, then state why the altered detail changes—or preserves—the legal comparison.

• Mark the changed feature.
• Name the fixed reference.
• State the recorded result.
• Reject one unfair comparison.
• Create one changed-detail retest.

4. Different totals, same percentage

Situation. A table shows 60% of 30 pupils and 60% of 45 pupils choosing an activity.

Tempting move. The numbers choosing are equal because the percentages match. This is attractive because it uses a visible detail quickly, but speed has replaced comparison design.

Diagnosis. The proportion is fixed but the whole changes, so the actual counts differ. Name the exact extra change, missing reference or role error. Avoid the blanket label “careless”; it cannot tell the learner what to notice next time.

Repair. Calculate 18 and 27 pupils; equal percentages do not imply equal amounts when totals differ. The repair keeps useful work and changes only the unsupported comparison or conclusion.

Check. Mark percentage as fixed and group size as changed. Read the answer backwards: result, comparison, changed feature and fixed reference. Each part should be recoverable from the task.

Transfer. Find totals that would make different percentages produce the same count. Solve the changed case without the model beside you, then state why the altered detail changes—or preserves—the legal comparison.

• Mark the changed feature.
• Name the fixed reference.
• State the recorded result.
• Reject one unfair comparison.
• Create one changed-detail retest.

5. Rectangles with fixed area

Situation. Several rectangles have the same area but different lengths and widths.

Tempting move. Compare perimeter by looking only at length. This is attractive because it uses a visible detail quickly, but speed has replaced comparison design.

Diagnosis. Width changes in response to the fixed area, so both dimensions determine perimeter. Name the exact extra change, missing reference or role error. Avoid the blanket label “careless”; it cannot tell the learner what to notice next time.

Repair. Use each complete length–width pair to calculate perimeter while keeping area as the stated fixed condition. The repair keeps useful work and changes only the unsupported comparison or conclusion.

Check. The conclusion is about shapes of equal area, not arbitrary rectangles. Read the answer backwards: result, comparison, changed feature and fixed reference. Each part should be recoverable from the task.

Transfer. Add a square and examine the pattern without claiming a general theorem beyond the data. Solve the changed case without the model beside you, then state why the altered detail changes—or preserves—the legal comparison.

• Mark the changed feature.
• Name the fixed reference.
• State the recorded result.
• Reject one unfair comparison.
• Create one changed-detail retest.

Previous chapter · Contents · Next: diagnose misconceptions

5. Diagnose misconceptions before adding volume

Repeated mistakes often come from a stable wrong rule. More questions can strengthen that rule if the learner never names it. Pause and classify the failure.

• comparing final totals when the starting totals differ
• holding the wrong quantity fixed
• assuming rows next to each other form a valid pair
• comparing averages without checking group size
• using a rate comparison when the time intervals differ

Turn each failure into a trigger–action pair. The trigger must be visible: “When two rows have different totals…” The action must be executable: “…I will convert to a common fraction or percentage before comparing.” “Be more careful” is neither visible nor executable.

Build contrast pairs. Place one legal comparison beside one illegal comparison that looks similar. Ask the learner to decide which pair can answer the question before doing any calculation or explanation. This trains selection rather than procedure-following.

Keep a compact error register with five fields: trigger, wrong move, why it fails, repaired rule and delayed retest date. Do not copy every line of the original task. Preserve only the information needed to retrieve the decision later.

Include correct answers that should not be changed. If every practice item contains an error, learners may begin to distrust sound work. Real checking requires a keep-or-repair decision.

Retest after attention has moved elsewhere. Immediate success may be memory of the correction. Delayed changed-detail success is better evidence of independent control.

Previous chapter · Contents · Next: worked repair set B

6. Worked repair set B

Reduce the support. For each case, write three short lines—changed, fixed, result—before reading the model. If a line cannot be filled, state what information is missing.

6. Fares across distances

Situation. A table lists taxi fares for journeys of different lengths and waiting times.

Tempting move. Subtract fares from rows with both distance and waiting time different. This is attractive because it uses a visible detail quickly, but speed has replaced comparison design.

Diagnosis. The price difference combines two changing causes. Name the exact extra change, missing reference or role error. Avoid the blanket label “careless”; it cannot tell the learner what to notice next time.

Repair. Choose rows with the same waiting time to compare distance, or the same distance to compare waiting. The repair keeps useful work and changes only the unsupported comparison or conclusion.

Check. A clean comparison isolates the relevant cost driver. Read the answer backwards: result, comparison, changed feature and fixed reference. Each part should be recoverable from the task.

Transfer. If no valid pair exists, state what additional row is needed. Solve the changed case without the model beside you, then state why the altered detail changes—or preserves—the legal comparison.

• Mark the changed feature.
• Name the fixed reference.
• State the recorded result.
• Reject one unfair comparison.
• Create one changed-detail retest.

7. Average scores

Situation. Two teams have averages of 8 and 9 points with different numbers of players.

Tempting move. The second team scored exactly one more point in total. This is attractive because it uses a visible detail quickly, but speed has replaced comparison design.

Diagnosis. The average difference is one point per player; total difference depends on team sizes. Name the exact extra change, missing reference or role error. Avoid the blanket label “careless”; it cannot tell the learner what to notice next time.

Repair. Compare average performance directly, or reconstruct totals only when group sizes are supplied. The repair keeps useful work and changes only the unsupported comparison or conclusion.

Check. Keep the unit ‘points per player’ attached to the comparison. Read the answer backwards: result, comparison, changed feature and fixed reference. Each part should be recoverable from the task.

Transfer. Make team sizes equal and show when total difference becomes easy to calculate. Solve the changed case without the model beside you, then state why the altered detail changes—or preserves—the legal comparison.

• Mark the changed feature.
• Name the fixed reference.
• State the recorded result.
• Reject one unfair comparison.
• Create one changed-detail retest.

8. Water tank readings

Situation. A table records tank volume at 9 a.m., 10 a.m. and 11 a.m., while water is added and used.

Tempting move. Compare net changes without noting different interval lengths. This is attractive because it uses a visible detail quickly, but speed has replaced comparison design.

Diagnosis. A larger change over a longer interval may not mean a faster change. Name the exact extra change, missing reference or role error. Avoid the blanket label “careless”; it cannot tell the learner what to notice next time.

Repair. Hold time interval fixed for a direct change comparison or calculate a rate when appropriate. The repair keeps useful work and changes only the unsupported comparison or conclusion.

Check. Label each interval before subtracting readings. Read the answer backwards: result, comparison, changed feature and fixed reference. Each part should be recoverable from the task.

Transfer. Create a two-hour interval and compare total change separately from hourly rate. Solve the changed case without the model beside you, then state why the altered detail changes—or preserves—the legal comparison.

• Mark the changed feature.
• Name the fixed reference.
• State the recorded result.
• Reject one unfair comparison.
• Create one changed-detail retest.

9. Recipe batches

Situation. Rows show flour, water and servings for different batches.

Tempting move. Compare water amounts while both flour and servings change. This is attractive because it uses a visible detail quickly, but speed has replaced comparison design.

Diagnosis. The result cannot identify whether the recipe concentration changed. Name the exact extra change, missing reference or role error. Avoid the blanket label “careless”; it cannot tell the learner what to notice next time.

Repair. Convert to the same batch basis or compare the water-to-flour ratio. The repair keeps useful work and changes only the unsupported comparison or conclusion.

Check. The fixed basis might be one serving or one part flour, depending on the question. Read the answer backwards: result, comparison, changed feature and fixed reference. Each part should be recoverable from the task.

Transfer. Double every quantity in one row and explain why the recipe ratio stays unchanged. Solve the changed case without the model beside you, then state why the altered detail changes—or preserves—the legal comparison.

• Mark the changed feature.
• Name the fixed reference.
• State the recorded result.
• Reject one unfair comparison.
• Create one changed-detail retest.

10. Discount offers

Situation. A table lists original price, percentage discount and final price for several items.

Tempting move. Choose the row with the biggest discount amount to identify the biggest percentage. This is attractive because it uses a visible detail quickly, but speed has replaced comparison design.

Diagnosis. Different original prices make amount and percentage separate comparisons. Name the exact extra change, missing reference or role error. Avoid the blanket label “careless”; it cannot tell the learner what to notice next time.

Repair. Hold original price fixed to compare discount amounts directly, or calculate each percentage from its own base. The repair keeps useful work and changes only the unsupported comparison or conclusion.

Check. Name whether the question asks dollars saved or proportion saved. Read the answer backwards: result, comparison, changed feature and fixed reference. Each part should be recoverable from the task.

Transfer. Construct two items with the same dollar discount but different percentage discounts. Solve the changed case without the model beside you, then state why the altered detail changes—or preserves—the legal comparison.

• Mark the changed feature.
• Name the fixed reference.
• State the recorded result.
• Reject one unfair comparison.
• Create one changed-detail retest.

Previous chapter · Contents · Next: checking language

7. Build a checking language that survives pressure

Use short questions: What am I comparing? What changed? What stayed fixed? Is the basis really common? What result was measured or stated? Does my conclusion name more than the result supports? These questions are faster than rereading with a vague feeling of doubt.

When a comparison is unfair, choose among three repairs. Select a better pair. Convert to a common basis. Or state that the available information cannot isolate the requested relationship. A mature answer sometimes recognises a limit instead of forcing a number or cause.

Check labels and units. A count, rate, percentage, average and total may use the same digits while representing different quantities. Attach the denominator or time interval until the comparison is complete.

For English, check speaker, claim and paragraph direction. For Mathematics, check whole, unit and interval. For Science, check variable roles, starting conditions and fair-test controls. The common audit stays stable while the subject-specific meaning changes.

Use a stopping rule. Stop when the comparison is fair, the result answers the task and every important claim has a visible source. Extra explanation should not introduce a second uncontrolled comparison.

Practise one checking question at a time, then mix them. Detection grows through focused attention; selection grows through mixed practice.

Previous chapter · Contents · Next: practice progression

8. Progress practice without random worksheet volume

Stage one is sorting: legal comparison, illegal comparison or insufficient information. Stage two is repair: fix the smallest failing part. Stage three is production: solve a fresh task. Stage four is mixed selection: decide whether this audit is needed among other skills. Stage five is timed transfer.

Advance by evidence, not page count. The learner should explain why a pair is fair and repeat the decision on changed details across more than one session.

Use near transfer first. Change one surface feature but preserve the relationship. Then use far transfer by moving from a table to prose, from prose to a diagram or from one subject to another. The learner must recognise structure without relying on topic labels.

Interleave. A page containing only “hold one variable fixed” questions announces the method. A mixed set forces the learner to retrieve the method when its trigger appears.

Delay one repaired item for two days, one for a week and one for a later mixed paper. The exact schedule can flex. The essential feature is return after forgetting has begun.

Track D–R–T: detected the issue, repaired it, transferred the rule. A correct answer without a visible route may be lucky. Transfer under changed conditions is stronger evidence of ownership.

Four audited examples can teach more than twenty rushed ones. The session ends when the learner can name the controlling difference and design a retest, not when the worksheet is empty.

Previous chapter · Contents · Next: worked repair set C

9. Worked repair set C

Hide the repair, solve independently and compare the structure of your answer—not only the final wording or number.

11. Bus capacity

Situation. Routes use buses with different capacities and numbers of trips.

Tempting move. Compare passengers carried by trip count alone. This is attractive because it uses a visible detail quickly, but speed has replaced comparison design.

Diagnosis. Capacity and trips both affect total possible passengers. Name the exact extra change, missing reference or role error. Avoid the blanket label “careless”; it cannot tell the learner what to notice next time.

Repair. Hold bus capacity fixed to study trips, or compute capacity × trips for total carrying potential. The repair keeps useful work and changes only the unsupported comparison or conclusion.

Check. The table must support the same occupancy assumption if actual passengers are compared. Read the answer backwards: result, comparison, changed feature and fixed reference. Each part should be recoverable from the task.

Transfer. Add partial occupancy and distinguish capacity from recorded use. Solve the changed case without the model beside you, then state why the altered detail changes—or preserves—the legal comparison.

• Mark the changed feature.
• Name the fixed reference.
• State the recorded result.
• Reject one unfair comparison.
• Create one changed-detail retest.

12. Fraction remaining

Situation. Rows show starting amounts, amounts used and amounts remaining.

Tempting move. Compare remaining amounts without checking starting amounts. This is attractive because it uses a visible detail quickly, but speed has replaced comparison design.

Diagnosis. A larger remainder can coexist with a larger fraction used if the starting whole differs. Name the exact extra change, missing reference or role error. Avoid the blanket label “careless”; it cannot tell the learner what to notice next time.

Repair. Hold the starting whole fixed or convert each remainder to a fraction of its own start. The repair keeps useful work and changes only the unsupported comparison or conclusion.

Check. The conclusion must name whether it concerns amount or proportion. Read the answer backwards: result, comparison, changed feature and fixed reference. Each part should be recoverable from the task.

Transfer. Create equal remainders from different starting amounts and compare fractions remaining. Solve the changed case without the model beside you, then state why the altered detail changes—or preserves—the legal comparison.

• Mark the changed feature.
• Name the fixed reference.
• State the recorded result.
• Reject one unfair comparison.
• Create one changed-detail retest.

13. Pattern table

Situation. Input–output rows appear to follow a rule.

Tempting move. Use two convenient rows and ignore one row that does not fit. This is attractive because it uses a visible detail quickly, but speed has replaced comparison design.

Diagnosis. A valid table rule must account for every stated pair unless an error or exception is explicitly given. Name the exact extra change, missing reference or role error. Avoid the blanket label “careless”; it cannot tell the learner what to notice next time.

Repair. Test the proposed operation on all rows and reject it at the first failure. The repair keeps useful work and changes only the unsupported comparison or conclusion.

Check. The fixed rule, not a visual trend, connects the rows. Read the answer backwards: result, comparison, changed feature and fixed reference. Each part should be recoverable from the task.

Transfer. Add another rule that fits two rows and use a third row to decide between them. Solve the changed case without the model beside you, then state why the altered detail changes—or preserves—the legal comparison.

• Mark the changed feature.
• Name the fixed reference.
• State the recorded result.
• Reject one unfair comparison.
• Create one changed-detail retest.

14. Survey years

Situation. A table compares survey results from two years with different sample sizes.

Tempting move. Subtract the counts and call it a change in preference. This is attractive because it uses a visible detail quickly, but speed has replaced comparison design.

Diagnosis. Count difference mixes sample-size change with preference change. Name the exact extra change, missing reference or role error. Avoid the blanket label “careless”; it cannot tell the learner what to notice next time.

Repair. Compare percentages when the question concerns relative preference, while noting sampling limits. The repair keeps useful work and changes only the unsupported comparison or conclusion.

Check. Hold the reporting basis at 100 respondents. Read the answer backwards: result, comparison, changed feature and fixed reference. Each part should be recoverable from the task.

Transfer. Make sample sizes equal and show how count and percentage comparisons align. Solve the changed case without the model beside you, then state why the altered detail changes—or preserves—the legal comparison.

• Mark the changed feature.
• Name the fixed reference.
• State the recorded result.
• Reject one unfair comparison.
• Create one changed-detail retest.

Previous chapter · Contents · Next: delayed transfer

10. Delay and transfer the skill

Transfer is the ability to recognise the comparison job when names, numbers, diagrams or contexts change. Begin with one-detail changes, then move the same decision into unfamiliar surfaces.

Use four passes. Solve without notes. Mark changed, fixed and result. Compare with the earlier rule. Finally, write the visible feature that triggered the rule. If the old model must remain open, independence has not yet arrived.

Create counterfactuals. What smallest change would make the rejected comparison fair? What change would break a valid comparison? Designing both versions reveals the boundary of the rule.

Mix correct and flawed work. Ask “keep, repair or cannot decide?” This prevents automatic editing and trains confidence in sound answers.

After a week, ask for a one-minute explanation without using the article title. The learner should give a fresh example and identify the baseline. This checks whether the idea is connected knowledge rather than a memorised phrase.

Transfer is strong enough for examination use when the learner detects the job, chooses a fair basis, computes or explains accurately, and stops without a teacher naming the method.

Previous chapter · Contents · Next: examination control

11. Control the examination answer

During a paper, use the full written audit for unfamiliar tables, multi-condition comparisons, investigation conclusions and purpose questions with several plausible roles. Use a mental version for routine cases.

Mark high-risk answers with a small symbol. Return to them first on the checking pass. This directs limited time toward decisions with more ways to fail.

If two comparisons compete, read the command word and name the requested quantity. “How many more,” “what percentage,” “which changed faster” and “what does the example show” require different fixed bases.

When time is short, secure the minimum complete answer: requested result, common basis and decisive evidence or unit. Add explanation only when the marks or wording require it.

Scan for comparison words, denominators, time intervals, starting values and causal verbs. These small features often decide whether generally correct work remains defensible.

Do not repeatedly rewrite. A controlled answer is one whose task, basis and conclusion match. Once that is true, move on.

Previous chapter · Contents · Next: independent audit

12. Run an independent audit

Choose one recent Mathematics response. Cover the correction. Copy the question and your answer. Label changed, fixed and result. Underline the strongest comparison or causal claim.

Decide whether to keep, convert, qualify, replace the pair or state insufficient information. Write the repaired answer beneath the original, preserving every correct part.

Add one sentence: “This comparison is legal because…” If the sentence cannot name a common basis, return to the source. If it relies only on the teacher’s authority, the rule has not yet been recovered.

Create a changed-detail retest for the next day. Alter the feature that controls fairness. Solve without the old answer. Record D, R and T.

When possible, ask another reader to identify the baseline from your answer. If they cannot, make the reference visible without adding unnecessary background.

Repeat weekly until the same failure stops returning under changed conditions. Then keep one later mixed retest to protect the skill from drift.

Previous chapter · Contents · Next: guidance and routes

13. Parent and tutor guidance, FAQs and a clear next route

What should an adult ask first?

Ask, “What are you comparing, and what is the common basis?” Let the learner point to the source before giving a method. Preserve correct work and repair the smallest missing relationship.

Should every answer name all controlled details?

No. Practice may show the full audit. Examination answers should match the task and marks. Include the decisive condition or reasoning bridge, not a catalogue of everything unchanged.

What if the learner knows the rule but forgets to use it?

That is a routing problem. Mix the target decision with other question types, remove topic labels and practise moderate time pressure. The learner must learn when to retrieve the rule.

How much practice is enough?

Use enough variation to prove that the decision travels. A short set with explanation, changed-detail retest and delayed return is more informative than a large same-day set.

What should a tutor record?

Record the missed trigger, unfair comparison, repaired rule and delayed-transfer result. Avoid the unexplained label careless. The record should tell the next session exactly what to test.

Where should the learner go next?

Return to the PSLE Learning Guide. Use the Mathematics hub for deeper canonical topic teaching. This volume owns the examination-performance decision, while detailed syllabus instruction remains with the established subject owners.

Continue through this four-volume batch

• Vol 0097: English: Separate the Writer’s Claim, Reason and Example Before Answering Purpose Questions
• Vol 0094: Mathematics: Hold One Quantity Fixed Before Comparing Rows in a Table
• Vol 0095: Science: Identify Manipulated, Responding and Controlled Variables Before Interpreting Results
• Vol 0096: English, Mathematics and Science: State What Changed, What Stayed Fixed and What the Result Shows

For the next session, audit one old answer, repair only the comparison boundary and create one changed-detail retest.

Previous chapter · Contents · Back to the start