PSLE MathsMisconceptionsSingapore

7 Most Common PSLE Maths Mistakes (And Why Kids Make Them)

Our bank classifies every wrong answer as misconception, calculation error or partial logic. Misconceptions win — here are the seven that cost the most marks.

Superholic Lab Editorial Desk8 min read

Where this data comes from

Superholic Lab's question bank holds 14,000 approved P5–P6 Mathematics questions, MOE-syllabus-aligned and written to PSLE format. Every multiple-choice question carries a pre-written explanation for each wrong option, classified into one of three failure modes: a misconception (a wrongly learned rule), a calculation error (right method, slip in arithmetic), or partial logic (started correctly, stopped a step early).

Across those questions, that is 16,962 classified wrong answers. The distribution surprised us:

  • Misconception — 7,393 (44%)
  • Calculation error — 6,517 (38%)
  • Partial logic — 3,052 (18%)

One honest caveat: this is a classification of how PSLE-style distractors are designed — the wrong answers a well-set question makes available to a student — not a statistic about how often real students pick each one. But that design is not arbitrary. Examiners build distractors from the mistakes markers actually see, which is why the same seven traps keep reappearing across two decades of papers.

The implication for parents is uncomfortable: “check your work” only addresses the 38%. Checking catches arithmetic slips. It cannot catch a misconception, because when the rule itself is wrong, the working looks perfectly correct to the child re-reading it.

1. "Of the remainder" is not "of the original"

A real question from our bank

Damian had $240. He spent 3/8 of it on a watch and 2/5 of the REMAINDER on shoes. How much did he spend on the shoes?

The trap answer: $96. That is 2/5 of the original $240. The question says 2/5 of the remainder — the money left after the watch. A child who has quietly learned the rule as “take the fraction of the big number” gets $96 confidently, every single time.

The actual answer: $60 — the watch takes 3/8 × $240 = $90, leaving $150; the shoes are 2/5 × $150 = $60.

The fix: Train one reflex: after every spending step, write down the new remainder before touching the next fraction. In bar-model terms — redraw the bar after each cut. The word “remainder” in a fraction question should feel like a flashing light.

2. Percentages do not add and subtract

A real question from our bank

A sum of money was increased by 50%, then the new amount was decreased by 40%. If the final amount was $360, what was the original sum?

The trap answer: $400 obtained the wrong way — or, more often, treating +50% then −40% as a net +10%. The child computes 50 − 40 = 10 and concludes the money grew by 10% overall. But percentages are multipliers, not counters: ×1.5 then ×0.6 is ×0.9 — the money actually SHRANK by 10%.

The actual answer: $400 — final = original × 1.5 × 0.6 = original × 0.9, so original = $360 ÷ 0.9.

The fix: Convert every percentage change into a multiplier before doing anything else (+50% → ×1.5, −40% → ×0.6), then multiply the chain. Never let two percentage signs touch each other with a plus or minus between them.

3. Reverse percentage: the discounted price is not 100%

A real question from our bank

During a mall promotion, a handbag is sold at a 20% discount. The discounted price is $96. What was the original price?

The trap answer: $115.20. That is $96 + 20% of $96 — adding the discount back onto the wrong base. The $96 is 80% of the original price, not 100% of anything. Discounts are taken off the original, so they must be undone from the original.

The actual answer: $120 — because 80% × original = $96, the original is $96 ÷ 0.8.

The fix: First sentence of working, always: “$96 is 80%.” Once the child writes down what percentage the known amount represents, the division is obvious. The trap only works on children who skip that sentence.

4. You cannot average two averages

A real question from our bank

In a P6 cohort, 8 pupils had an average Science score of 75 marks. Another 2 pupils joined and scored 85 marks each. What is the new average of all 10 pupils?

The trap answer: 80 marks. That is (75 + 85) ÷ 2 — averaging the two averages as if each group counted equally. But one “average” represents 8 pupils and the other represents 2. Averages carry different weights, and adding them head-to-head throws the weights away.

The actual answer: 77 marks — total = 8 × 75 + 2 × 85 = 770, divided by 10 pupils.

The fix: Drill the identity “average × number = total” until it is automatic. Every average question is secretly a totals question: convert to totals, combine, then divide once at the end.

5. Dividing by a fraction makes the answer bigger

A real question from our bank

A fruit-juice stall has 6 litres of watermelon juice. Each serving is 3/4 of a litre. How many complete servings can be poured?

The trap answer: . That is 6 × 3/4 — the child multiplied because “division makes things smaller” felt wrong, so they quietly swapped the operation. The intuition that division shrinks numbers is built in P2–P3 on whole numbers, and it breaks the first time the divisor is smaller than 1.

The actual answer: 8 — 6 ÷ 3/4 = 6 × 4/3 = 8 full servings.

The fix: Reframe division as “how many of these fit inside?” How many three-quarter-litre cups fit in 6 litres? Clearly more than 6. When the divisor is less than 1, the answer MUST be bigger than the starting number — teach that as a sanity check, not a surprise.

6. Ratio change: find what stays the same first

A real question from our bank

Mr Lee mixed blue and white paint in the ratio 2 : 7, using 28 litres of white paint. He wants to add more white paint so that the new ratio of blue to white becomes 1 : 5. How many more litres must he add?

The trap answer: Anything computed by re-scaling both quantities. The child treats the new ratio 1 : 5 as a fresh problem and re-computes both paints, forgetting that only white paint was added — the 8 litres of blue never changed. Before–after ratio problems collapse instantly once you anchor on the unchanged quantity; they are nearly unsolvable if you do not.

The actual answer: 12 litres — blue stays at 8 L, so new white = 5 × 8 = 40 L, which is 12 L more than 28 L.

The fix: First question on every before–after ratio problem: “what did NOT change?” Circle it, compute its actual value, and rebuild the new ratio around it. This single habit converts the hardest P6 ratio questions into two-step arithmetic.

7. Angles at a point make a full turn — 360°, not 180°

A real question from our bank

At a point, angles are 90°, e°, 2e°, and 30°. Find e.

The trap answer: e = 20°. That comes from setting the sum to 180° — mixing up “angles at a point” with “angles on a straight line”. The two rules are taught weeks apart in P5 and merge into one blurry memory by exam season.

The actual answer: e = 80° — since 90 + e + 2e + 30 = 360, we get 3e = 240.

The fix: Have your child physically turn on the spot: a full turn is 360°, a half turn (straight line) is 180°. Then make them write the rule they are using — “angles at a point = 360°” — as the first line of working. Naming the rule prevents the swap.

What to do with this as a parent

The pattern across all seven: each misconception produces the same wrong answer on every question built the same way. That consistency is your diagnostic. Random wrong answers are carelessness; consistent wrong answers are a wrongly learned rule — and a rule can be re-taught in an afternoon once it has been named.

So when the same mistake appears for the third time, resist “you were careless again”. Instead, work backwards from the wrong answer and ask: what rule would make this answer feel right? Then fix that one rule, on its own, before mixing the topic back into general revision.

This is, transparently, the entire design premise of Superholic Lab: every wrong answer a child picks is met with the explanation for that specific misconception, our diagnosis layer finds the root-cause topic behind repeated mistakes, and a 3-day Plan Quest fixes one rule at a time. But the method works with nothing more than a pencil and your child's last three test papers — start there tonight.

Want to see which of these traps your child falls for? The free topic quizzes cover every topic above, no sign-up needed — and the common mistakes library lists the misconceptions for every P1–P6 topic.

Frequently asked questions

Are these mistakes just carelessness?

Usually not — and there is a simple test. A careless slip is random: different wrong answers on similar questions. A misconception is consistent: the same wrong answer, every time, on every question built the same way. If your child confidently answers $96 on every "of the remainder" question, that is a wrongly learned rule, not a lapse in attention — and telling them to "be more careful" cannot fix it.

Which of these misconceptions costs the most marks in the PSLE?

Percentage and fraction misconceptions are the highest-stakes because they appear in both Booklet A (1–2 marks each) and the heavily weighted Paper 2 word problems (3–5 marks each). A single unfixed misconception — say, taking the fraction of the original instead of the remainder — can silently repeat across several questions in one paper.

How do I find out which misconception my child actually has?

Collect three to five of their recent wrong answers on the same topic and look for a pattern: is it the same kind of wrong answer? Working backwards from the wrong answer usually reveals the rule the child actually used. This is exactly what Superholic Lab automates — every wrong option in our question bank carries a pre-written explanation of the specific misconception that produces it, so the diagnosis happens on every practice question.

At what level should these misconceptions be fixed?

Most of them take root in P4–P5 when the underlying topic is first taught, and surface at P6 under exam pressure. The earlier the fix, the cheaper it is — a P5 child has a full year of practice to consolidate the corrected rule, while a P6 child discovered in September has weeks. That said, none of these is hard to fix once it has been named: each is one rule, not a whole topic.