A simple, repeatable method for working out why your child got a question wrong — not just what the right answer was.
25 August 2026 · 5 min read
When a child gets a maths question wrong, the instinct is to show them the right answer and move on. That fixes the one question in front of you. It usually doesn't fix the reason it happened, which means the same mistake tends to resurface a week later on a different worksheet.
Finding the actual root cause doesn't require any special training — just a different set of questions, asked in the right order. Here's the method.
"What's the correct answer?" invites a child to guess again or wait for you to tell them. "Can you show me how you got that?" invites them to walk you through their actual thinking — and that's where the mistake usually reveals itself, long before you get to a final number.
Take a fairly common example: a child subtracting 502 − 378 and getting 276. Marked purely on the final answer, that's just "wrong." But look at the working and a pattern often shows up — in each column, they've subtracted the smaller digit from the larger one regardless of which number it belonged to (8 − 2 = 6, then 7 − 0 = 7, then 5 − 3 = 2), rather than borrowing correctly. That's not a random error. It's a well-known, specific misconception about how borrowing works, and it will produce a wrong answer on almost every subtraction question that needs a borrow.
Once you have a theory, check it. Give a subtraction question that doesn't require any borrowing at all — if that's correct, ordinary subtraction isn't the issue. Then give one that requires borrowing in exactly one column. If the same "smaller from larger" pattern shows up again, you've confirmed it: the gap is specifically in borrowing, not subtraction generally, and not this particular question.
This step matters because it's easy to over-correct. If you only saw one wrong answer, it might have been a slip, not a gap — testing the pattern once more is what tells the two apart. We wrote more about that distinction in how adaptive learning identifies maths gaps.
If your child solved a similar-looking question correctly last week, put the two side by side and ask what's different. Often the honest answer is genuinely useful: "that one didn't need borrowing" or "I knew that one off by heart." That comparison usually narrows the gap down faster than analysing the wrong question alone.
If the pattern holds, step back to the skill underneath it rather than drilling more questions at the same level. For borrowing, that might mean a quick check on place value first — does the child actually understand that the "5" in 502 represents 500, not 5? A gap in borrowing sometimes isn't about the subtraction procedure at all; it's a place value gap wearing a subtraction costume. Getting that one level lower is usually where the real root cause is sitting.
Put together, the method is really five short questions, in order:
It takes a few minutes, and it usually gets further than a whole extra worksheet on the same topic.
This is exactly the process a genuinely adaptive system automates: checking a wrong answer against known misconception patterns, testing whether it's a one-off or a pattern across several questions, and stepping back to the prerequisite skill when a gap is confirmed — the same "backtrack one level" logic, just running in the background on every question, not only the ones you happen to catch. That's the scaffolding we described in what is adaptive learning in maths.
Grasp Maths runs this kind of root-cause check on every wrong answer, not just the ones a parent happens to have time to sit down and investigate — so the next question a child sees is aimed at the actual gap, not just another version of the one they got wrong.
Don't just practise more. Practise what matters next.
Try a free Grasp Maths practice session and see how the engine traces a mistake back to its root cause automatically.