From P2 to P3 Maths: What Actually Changes

26 August 2026 · by Larry

A child comes home in the first term of Primary 3 with a maths paper that is worse than anything they brought home last year. Nothing about them has changed. They are not lazier, they have not stopped trying, and if you sit them down and ask them to add or subtract, they still can.

What changed is the shape of the questions, and it changed in four ways at once. That is why the drop can feel so sudden, and it is also why "do more practice" often does not fix it — it is usually the wrong medicine for the actual problem.

1. There is much more to read

The single biggest difference between a P2 paper and a P3 paper is the amount of English in it. Questions get longer. They contain more information, and some of that information is not needed. Names, objects and two or three numbers arrive before the actual question does.

A child who can do the arithmetic perfectly can lose the mark because they took the wrong number out of the sentence, or answered the question they expected instead of the one that was asked.

This is worth being precise about, because it is not a maths problem and treating it as one wastes everyone's evening. More sums will not help a child who is misreading.

2. The questions became multi-step

In P2, most questions ask for one thing. You find the operation, you do it, you write the answer. A child can get a long way by recognising the type of question.

In P3 the answer to the first part becomes the input to the second. You cannot spot your way through it; you have to hold a plan in your head. First I need to know how many she had left, and then I can share them out.

This is a genuinely new skill and it is the one most children are actually missing when their marks dip. It looks like weak maths and it is really weak planning.

3. The numbers got bigger, so old methods stop coping

Strategies that were fine last year quietly stop working. A child who still works out multiplication by adding repeatedly, or who counts to get to small sums, was managing perfectly well when the numbers were small. Now the same method takes so long that they lose the thread of the question while doing it.

The arithmetic is not beyond them. It is just using up all the attention, leaving none for the actual problem. This is the point at which the times tables become load-bearing rather than a nice extra.

4. Method starts to earn marks

Working now has to be visible. A correct answer with nothing written down can still lose marks, and a wrong answer with sound working can still gain some. Children who have spent two years being rewarded for the answer alone find this baffling and slightly unfair, and they resist it.

It is also where marks disappear from a paper that looks correct.

How to tell which one you are dealing with

Take one question they got wrong. Do not help, do not explain, and do not let them solve it. Ask them to read it aloud and then tell you, in their own words, what it is asking them to find.

Three things can happen, and each one points somewhere different:

• They cannot say what is being asked. It is a reading problem. The maths is fine.
• They say it correctly but cannot say what to do first. It is a planning problem — the multi-step skill above.
• They say it, they know the first step, and the arithmetic goes wrong. Only now is it actually a maths problem.

Those three need three different responses, and doing the wrong one is why a term of extra worksheets can produce no improvement at all.

What helps, in order

Ask what they are being asked to find, before any calculation. Every time. Have them underline it in the question. It takes a few seconds and it prevents the most common way a mark is lost.

Practise first steps only. Give a question and ask only for step one — then stop them. Do not let them finish. It feels strange, and it is the fastest way to build planning, because it separates the thinking from the arithmetic that usually swallows it.

Draw it. When the words are in the way, a picture usually gets past them. This is exactly what bar models are for, and it is why they arrive around this stage.

Get the tables solid. Not because tables are the point, but because everything above sits on top of them. A child spending attention on six times seven has none left for the question.

Insist on written working even when they can do it in their head. Especially then. The habit has to exist before the questions get hard enough to need it.

Do not add volume. Ten more questions of the same kind will not teach planning. One question worked through slowly, with the first step said out loud, is worth more than a page.

Three things not to do

• Do not time them yet. Speed comes after the method is reliable, not before. Pushing pace while the method is still forming produces exactly the kind of rushed mistakes that look like carelessness.
• Do not correct mid-question. Let them finish the wrong answer. You learn far more from where it went wrong than from a question you steered.
• Do not conclude anything from term one. A dip at this exact point usually says more about a change in how questions are asked than about what a child can do.

The thing worth saying out loud

Children draw conclusions from marks quickly, and a run of worse papers turns into a sentence about themselves surprisingly fast. If you hear the beginnings of "I'm bad at maths", it is worth naming what actually happened: the questions changed, they have not caught up yet, and that is a completely different thing from not being able to do it.

That is true, it is more useful than reassurance, and it happens to be the explanation that matches the evidence in front of you.

A step up like this is one of the ordinary explanations for a mark that falls sharply — see what to check first when maths marks drop suddenly.

There is a second jump two years later, and it catches a different child — what changes between P4 and P5.