Maths marks are lost in four distinct ways: the concept was never understood, the concept is understood but the method is wrong, the method is right but execution slips, or the question was never attempted. Only the first two are taught away. The other two need different fixes, and paying for teaching to solve them is why some families see no change after a year.
Why maths breaks differently from other subjects
Maths is cumulative in a way most subjects are not. A child who missed a fortnight of history has a gap in history. A child who did not secure fractions in P4 will struggle with ratio in P5, algebra in Sec 1, and rational functions in Sec 3 — each new topic silently assuming the earlier one held.
This is why “he was fine last year and suddenly he’s failing” is so common a report, and why it is usually inaccurate. The failure is rarely sudden. It is the point at which an old gap finally became load-bearing.
It also means the useful question is not “what is he failing now” but “where did it stop making sense”. Those are different questions with different answers, and the second one is what a good diagnosis finds.
The four gaps, and what each needs
Conceptual gap — the idea was never built. The child can follow a worked example step by step but cannot start an unfamiliar question. Asked why a step works, they say “that’s the formula”. This is the deepest gap and the most expensive to leave: every later topic built on it inherits the weakness.
It needs teaching that starts before the formula. A child who understands that dividing by a fraction asks “how many of these fit into that” will handle it in any disguise. A child who memorised “flip and multiply” is fine until the question is phrased unusually.
Procedural gap — the idea is there, the method is not. The child understands what is being asked but chooses an approach that cannot get there, or knows two methods and picks the wrong one. This responds well and quickly to teaching: exposure to worked variety, and practice at recognising which question type they are looking at.
This is the gap where tuition delivers the fastest visible improvement, which is also why it gets over-diagnosed. It is genuinely common — but so is assuming it when the real problem is conceptual.
Execution gap — right idea, lost marks. Sign errors, transcription slips, working that skips the step the marker needed, answers not in the form asked for. The child knows the maths and the paper does not show it.
More teaching does not fix this. Consistent, strict marking does — someone who takes the mark off every time the working is incomplete, until the habit changes. That can be a tutor, but it can also be a parent with the marking scheme.
Attempt gap — nothing written. Blank questions, papers abandoned in the last section. Usually time management, sometimes a decision that the question is beyond them before reading it properly. This needs timed practice and a rule about attempting every question, not more content.
How to run the diagnosis at home
Take the last two marked papers. For each lost mark, ask one question: could my child have got this if they had more time and no pressure?
Yes, easily → execution or attempt gap. Yes, but slowly → procedural gap; the method is there but not fluent. No, they would not know where to start → conceptual gap.
Then ask them to explain one wrong answer aloud, without writing. Children reveal the difference immediately. A procedural gap sounds like “I did it this way but I think I should have done the other one”. A conceptual gap sounds like restating the question in different words, or silence.
This takes twenty minutes and is more useful than any placement test, because you are watching your own child think.
What good maths teaching does about each
Once you know which gap you are dealing with, you can judge whether a tutor or centre is addressing it — which is a much better question than whether they are “good”.
For a conceptual gap, watch whether the teaching starts from meaning or from procedure. Does the tutor ask the child to explain what the question is asking before doing anything? Do they use a diagram, a concrete example, a number line? Or do they demonstrate the method and set practice?
For a procedural gap, watch for variety. Is the child seeing the same question type in different disguises, so recognition builds? Or is it the same shape repeated, which builds speed on a shape that will not appear that way in the exam?
For an execution gap, ask to see marked work. Is the marking strict about presentation? Does it comment on working, or only on the final answer? Loose marking actively teaches a child that incomplete working is acceptable.
For an attempt gap, ask what the tutor does about pacing and about a child who gives up. If the answer is vague, the gap will not close.
The Singapore-specific part
Two features of the local syllabus are worth understanding because they shape where children get stuck.
Model drawing and heuristics at primary level teach a way of representing a problem before solving it. Children who skipped this and moved straight to algebra-like shortcuts often cope until the word problems get complex, then lose the ability to represent what is happening.
The jump at Secondary 3 is real, particularly into Additional Mathematics. The pace increases, the abstraction increases, and the assumption that Sec 1–2 algebra is fluent is absolute. A child who was managing with partial fluency usually stops managing here.
Neither of these is a reason to panic. Both are reasons to diagnose early rather than wait for a result that confirms it.
Frequently asked questions
How do I know if my child needs maths tuition at all? If the lost marks are conceptual or procedural and the school cannot give individual attention to it, tuition addresses the problem directly. If the marks are lost to carelessness or unfinished papers, try strict marking and timed practice first — they are cheaper and often sufficient.
My child understands in class but fails tests. What is that? Usually a procedural or execution gap. Understanding while watching someone else work is a much lower bar than producing the work alone under time. Ask them to do a question with you watching but not helping — the difference shows quickly.
Should we go back to an earlier topic? If the gap is conceptual, yes, and it is the fastest route even though it feels like going backwards. Repairing ratio in P5 is quicker than fighting every ratio-dependent topic afterwards.
How long should it take to see change? Procedural and execution gaps often show movement within a term. Conceptual gaps take longer because something is being rebuilt rather than corrected. If nothing has moved in a term with good attendance, the diagnosis was probably wrong — go back to the four gaps.
Group or one-to-one for maths? One-to-one is better for diagnosis and for conceptual repair, because it can go at the child’s pace and follow their reasoning. Group is efficient for procedural practice and pacing once the concept is secure. Many families use the first briefly, then the second.
—