In maths, revisiting mistakes often helps more than doing pages of practice

A student can spend an hour filling pages of maths and still make the same sign errors, fraction mistakes or wrong method choices in the next test. For a parent, that is baffling: the work has clearly been done, it is visible, sometimes lengthy, and yet the result does not really follow.

In many cases, the explanation is simple. In maths, reworking mistakes precisely is often more useful than doing large amounts of additional practice. Progress comes less from raw quantity than from the ability to find the exact line where the reasoning broke down, understand why, and then check a few days later that the same error has not returned.

This is not an argument against practice. It is an argument against blind practice. Pages become useful again once the real obstacle has been identified. Before that, they can reassure more than they teach.

Why this strategy often saves time

In maths, a failed question is not just a failure. It is a diagnosis. It shows whether the student has not understood the idea, knows the rule but applies it at the wrong moment, skips a step, or simply does not recognise the type of problem in front of them. Reworking that question properly forces them back to the precise point that gave way.

By contrast, doing twenty almost identical questions can create an illusion of work. The student ends up imitating a model, recognising the shape of the page, or following a procedure by reflex without really choosing it. That may be enough in a very guided session, but the transfer to a new test remains fragile.

Time is mostly lost in three situations:

• the solution is copied, but not reconstructed;
• the whole worksheet is blocked by one repeated micro-obstacle;
• tiredness turns the session into mechanical repetition.

So the real problem is not always a lack of practice questions. Very often, it is the lack of a precise diagnosis. In maths, the real unit of progress is not the completed page. It is the mistake that stops coming back.

Identify the real difficulty

Before asking for more work, it helps to name the kind of difficulty involved. This table gives a simple guide and avoids some very common false remedies.

This diagnosis changes a great deal. You do not deal with weak basic fluency by giving a long explanation of the lesson, and you do not fix a deep misunderstanding with twenty questions copied from the same model.

In the earlier secondary years, blockages often come from basics that absorb all the student's attention: signs, fractions, order of operations, moving from words to calculation. In the later secondary years, the difficulty more often shifts towards method choice, written justification, or recognising a less familiar problem. In early higher education, older gaps become more costly because the pace no longer allows everything to be rebuilt at the last minute.

The useful question is therefore not Have you understood? It is: At which line did you stop knowing what to do?

A simple method for reworking a failed question

Going back over mistakes only helps if the correction becomes active learning. A simple method is often enough.

1. Try the question again without looking at the correction. Even if the student does not finish, this attempt shows what they can still do on their own.
2. Find the first wrong line. There is no need to comment on the whole page in one block. The aim is to locate the exact point where the reasoning starts to drift.
3. Name the error in one short sentence. For example: I distributed the minus sign too quickly; I chose the right formula but not the right theorem; I did not notice the unknown was on both sides.
4. Review only the missing piece of method. When the idea is new or still unclear, one worked example step by step often helps more than another whole page of questions.
5. Check with twin questions. Two or three very similar questions are enough to see whether the correction really holds.
6. Come back to the same point later. A short return the next day or two days later is usually more informative than one long session done all at once.

A concrete example

Suppose a student writes -3(x - 2) = -3x - 2. The problem is not equations in general. The real breakdown is the distribution of a negative factor. So the right work is not to redo the whole chapter. It is to rework that line, then handle a few very similar expressions with brackets, and only then return to an equation where that step really matters.

In this kind of situation, a parent does not need to reteach the whole topic. What they can insist on is precision.

Three useful questions for a parent

• Where does the error start exactly?
• Which rule or method choice was missing?
• Which almost identical question will let us check tomorrow that this has really been corrected?

Even with very little time, those three questions help more than doing the correction in the student's place.

An error log can also help, provided it stays light. One line per error is enough: type of error, rule to remember, twin question, date of the next check. Otherwise, the notebook quickly becomes an archive of failures rather than a tool for progress.

When more practice becomes useful again

The important word in the title is often, not always. Once the error has been identified, more questions can become very useful again. But the aim is then different.

To automate a move that is already understood. When the method is clear but still slow or fragile, a modest amount of repetitive practice can strengthen the basics. This often happens with algebraic manipulation, fractions, common algebraic identities, or some routine differentiation techniques.

To learn how to choose the right strategy. When the problem is no longer execution but recognition of the right tool, mixed sets are more useful than uniform blocks. They force the student to ask, before calculating, what kind of problem this is.

To prepare for an assessment format. A timed test also requires time management, question order and sustained attention. At that point, a longer practice session can make sense, provided understanding is already there.

To consolidate over time. A topic does not become stable because the student got it right once on the evening of the correction. It becomes stronger when it is seen again later, first in a similar form and then in a slightly different one.

In practice, that leads to a simple rule. When the student is lost, first guide and correct precisely. When they are beginning to master the method, then vary, space and make the recall more independent. Doing whole pages is not useless in itself; it is simply a poor first reflex when nobody yet knows what is really blocking progress.

How to tell whether progress is real

The best indicator is not the number of minutes spent or the number of pages filled. It is the quality of transfer. A student is really progressing when several signs appear together:

• they can explain why they are choosing a method before starting the calculation;
• the same error disappears on a similar question and then on a mixed one;
• they need less outside help to get started;
• their written work becomes clearer and more stable, with fewer backwards steps;
• a few days later, they can still find the right procedure without rereading the whole correction.

By contrast, some signals suggest that more structured support may be needed. That is the case when the same basic weakness runs through several topics, when the student still cannot explain a correction that has already been revisited several times, or when working time increases while avoidance and discouragement increase as well. In such cases, a conversation with the teacher can help distinguish a fragile method from an older gap or a more specific support need.

Before saying Do the whole sheet again, three questions are often enough:

1. What is the exact step that is blocking?
2. Is the right next move a guided example, a twin question or a small mixed set?
3. When will we check that this error has not come back?

In maths, visible progress is not the pile of completed sheets. It is the number of mistakes that stop returning.