Maths Anxiety in the Classroom: Four Ways Adults Respond, and What the Evidence Says About Each

Quick answers

Start here, because it reverses the usual assumption

The ordinary reading is that a child is anxious about maths because they are weak at it, so strengthening the maths will dissolve the anxiety. One of the sources we read — a review written for teachers — reports a finding that cuts against this: children with the most working memory show the strongest negative relation between anxiety and achievement.

The explanation it offers makes the result stop feeling strange. High-working-memory students tend to use more demanding strategies: holding intermediate steps in mind, trying a route and backing out of it, checking an answer against an estimate. Those strategies are why they do well, and they are also the most sensitive to interference, because they need exactly the mental space that worry occupies. A student who counts on fingers is slower but harder to knock off course.

The classroom consequence is immediate. When a normally strong student’s test score drops while their homework stays excellent, the explanations that come first — careless, unmotivated, did not revise — may all be wrong, and all three lead to the same unhelpful prescription: more practice on material already mastered. That raises the strong side and leaves the obstacle untouched.

The four responses, ranked by what the evidence supports

We read a peer-reviewed meta-analysis pooling 50 studies and 75 effect sizes for anxiety (about 9,125 participants), plus 41 effect sizes for performance (about 6,097 participants). It grouped interventions into three families. Set beside the review for teachers, four practical responses fall out:

What we are not saying. Not that encouragement is harmful, and not that warmth has no value of its own. Two things follow and no more: it has not been shown to reduce maths anxiety, and stopping there leaves the two families that did work unused. The review also cautions that the motivation null may reflect too few studies (only 15 effect sizes) rather than a genuine absence of effect — and we publish that caution rather than leaning on the result harder than it will bear.

What to say instead, when a child is stuck right now

The teacher-facing source sets out a three-part alternative to consoling, and it is specific enough to use verbatim:

1. Acknowledge that the problem is genuinely hard. Not “it’s easy, look” — which tells a stuck child something false about themselves.
2. Express confidence that they will get there. About this problem, with this amount of work.
3. Hand them a concrete strategy. One next move. Not a feeling, a step.

The difference between this and reassurance is that steps one and three carry information. “You’re so clever, don’t worry” carries none, and the child already knows the answer is not on the page.

🔴 The part that includes us — adults transmit this

The source reports a preliminary study of 17 teachers and 117 students in which “the higher a teacher’s math anxiety, the lower her female students’ math achievement” by year’s end, and a follow-up with more than 70 teachers and 650 students finding the teacher’s anxiety carried implications for boys as well, though less strongly.

It also reports something useful about fixing it: for trainee teachers, a course on how to teach maths was more effective at reducing their own maths anxiety than a course on the maths content itself. Confidence came from knowing what to do in the room, not from more content.

This is a point about situations, not a reason for blame. The practical form it takes is small: “I was never a maths person either” is said as solidarity and lands as permission. Say the subject is hard, and that hard is normal, instead.

Our own position, stated plainly

Three caveats we publish rather than bury

• The authors say their own numbers probably run high. A publication-bias test showed asymmetry, and they write that “potentially the present results overestimate the effects of the interventions for the math anxiety part.”
• Study quality was uneven. They state that “the methodological quality of many studies included in our meta-analysis was generally low,” and that about 16% used a single-cohort pre-post design with no control group. A study without a control group cannot separate the intervention’s effect from the passage of time, the child’s own development, and a new adult paying them attention — and the last of those is itself a strong intervention.
• The performance side is weaker than the anxiety side. The authors note the performance analysis “might be considered to suffer from a power-issue.” Reducing anxiety and raising marks are two claims, and the second rests on less.

And the limit that applies to all of it: these are group-level averages. They tell you which bets are better, not what your child is. Your child is not a group.

When you do not need to do anything — or not yet

1. You have not yet tried removing time pressure at home. It is the cheapest change available and it is unreasonable to buy a solution before testing the free one.
2. The score is low and the child is calm. That pattern points at the maths, not the anxiety, and treating anxiety that is not there spends effort for nothing.
3. The tension is the same across every subject. Then the subject is not the variable, and a maths-specific response will not reach it.
4. The child is under 12 and you are considering a paid anxiety programme. The age moderator in the meta-analysis does not support that bet at that age; situational change and foundational number work are the better use of the same money.
5. The gap is one named topic. One or two targeted sessions on that topic beat a general course, and cost less.
6. The real problem is sleep or an overloaded week. Another lesson on top of a full timetable adds load, not learning.

If it does turn out that the maths itself is the obstacle, you can message us to locate where the obstacle sits before committing to anything, or read how online support for learning differences works.

Frequently asked questions

Does maths anxiety mean my child is weak at maths?

Not necessarily. The teacher-facing source reports that students with the highest working memory show the strongest negative relation between anxiety and achievement, because their strategies are the most demanding and so the most disrupted. It also notes that weaker early building blocks raise the risk of developing anxiety, so the relationship runs in both directions rather than from a single cause.

What is the fastest change I can make?

Remove timers and speed competitions from home practice, and let the child write steps on scrap paper. The first because the source reports the link to poor performance is stronger on timed tests, the second because it offloads working memory, which is the mechanism the significant intervention category targets.

Isn't encouragement the right response?

In the meta-analysis we read, the motivation category was the only one of three that did not reach statistical significance. That does not make encouragement harmful; it means it has not been shown to reduce maths anxiety, and that stopping there leaves two effective categories unused. The authors also caution the null may reflect too few studies in that category.

Should I help when my child is stuck, or let them struggle?

The source warns that consoling a struggling student can validate the belief that they are bad at maths, and offers a three-part alternative: acknowledge the problem is hard, express confidence they will get there, then give a concrete next step rather than reassurance.

Does writing about feelings before a test help?

The source reports that about 10 minutes of writing about one's feelings before an upcoming maths test helped reduce the performance gap between higher- and lower-anxiety students. It was studied in older students, so extending it to a young child is an inference rather than something the source states.

Could I be part of the problem?

The source reports two studies linking a teacher's own maths anxiety to lower student achievement, in girls first and in boys to a lesser degree. It is a point about situations rather than blame, and the useful consequence is to watch what you say about the subject in front of the child.

My child is in primary school. What is most likely to help?

The meta-analysis found interventions had a small effect at age 12 or under and a moderate effect above 12, a statistically significant difference. At younger ages the better bets are changing the situation and strengthening foundational number and spatial skills, rather than an anxiety-targeted programme.

When should I involve a professional?

If the distress is broad rather than specific to maths, persists, or disrupts the child's sleep, eating or daily life. This is an educational page: we do not diagnose or treat.

Sources and caveats

Read on 14 October 2026: (1) a peer-reviewed meta-analysis of maths anxiety interventions — 50 studies, 75 effect sizes for anxiety (9,125 participants) and 41 for performance (6,097 participants), with intervention categories, age and length moderators, and the authors' own limitations; (2) a review article written for teachers and published in a professional education periodical, summarising the cognitive and classroom research on maths anxiety.

All figures above are group-level. The intervention review reports publication-bias asymmetry and generally low methodological quality among included studies, and says its own results may overestimate the effects. Maths anxiety as used here is an educational description, not a clinical diagnosis, and nothing on this page is an assessment of an individual child.