Math Anxiety: Why Your Child Freezes on Tests, and What Helps

What math anxiety actually is

The definition most researchers still use dates to 1972, from Richardson and Suinn. The Centre for Neuroscience in Education at Cambridge carries it: "a feeling of tension and anxiety that interferes with the manipulation of numbers and the solving of mathematical problems in ... ordinary life and academic situations."

The verb doing the work in that sentence is interferes. Something is already there, and the anxiety gets in front of it.

Cambridge is blunt about who this happens to: "otherwise perfectly intelligent and capable persons develop a severe avoidance of situations involving any kinds of mathematics." That is worth sitting with for a moment if you have been quietly wondering whether your child is simply not a math person.

Be precise about what that does and does not mean, though. Ashcraft's finding is that highly math-anxious people "do not have a global deficit in math competence." It is not that competence is untouched. He is equally clear that avoidance "ultimately undercuts their math competence", because anxious students take fewer math courses and "apparently learn less of what they are exposed to." Ability is not the starting problem. Given enough years of avoidance, it can become one.

How common is it? The honest answer is that nobody has published a Canadian figure. The nearest solid number comes from the UK. The Centre for Neuroscience in Education puts the prevalence of extreme math anxiety at "between 2-6% at secondary school level in the UK", citing work by Chinn. Extreme is the operative word there. Milder versions are more common, and Cambridge notes those "can still have a significant effect on the people who suffer with it."

It shows up in the body. Ashcraft, who has spent a career measuring this, opens his 2002 review with an account from his own lab: a participant grew steadily more uncomfortable during a testing session until she "burst into tears." His research assistant reported that many participants show "trembling hands, nervous laughter, and so forth", and Ashcraft notes these reactions "are not triggered by deliberately provocative procedures." Nobody was being unkind to her. The task was to "solve elementary-school arithmetic problems", under time pressure.

One caveat to carry through the rest of this piece. Those participants were "college adults". Most of the precise laboratory work on this mechanism was done on university students rather than on children. The mechanism is about how working memory behaves under worry, which is not a specifically adult property, but the measurements are adult measurements and I will not pretend otherwise.

So a child who says their stomach hurts before a math test may well be reporting something real. I would not treat that as a bid to stay home.

Now a distinction that matters more than it first appears. Cambridge is explicit that math anxiety "is also not to be confused with general anxiety related to assessment. That is, Mathematics Anxiety is anxiety related specifically to mathematics, not anxiety about taking tests." The exam is usually where a parent first notices the problem. The problem itself is attached to the math, and Cambridge adds that it "is not restricted to tests or classroom settings." A child can be perfectly composed in a history exam and come apart over a worksheet at home.

Now the caveat most pages skip. Higher math anxiety and weaker math performance do travel together, and that association has been measured many times. Which one causes which is not settled. Cambridge puts it as a chicken-and-egg problem and says plainly, "What we don't know is which causes which." A separate question, where math anxiety comes from in the first place, is also open: Ashcraft writes that there has been "no thorough empirical work on the origins or causes of math anxiety." Anxiety can crowd out performance, weak performance can breed anxiety, and in a real child both are usually running at once. Anyone who tells you the arrow points one way is going beyond the evidence.

Why a child who knows the method still freezes

This is the part almost nobody explains to parents, and it is the whole reason freezing looks so much like forgetting.

Working memory is the small workspace your mind uses to hold a few things at once while you do something with them. Carrying a digit. Keeping the first half of a word problem in mind while you read the second. It is genuinely limited, for everyone.

Ashcraft's summary of the research is one sentence long: "Math anxiety disrupts cognitive processing by compromising ongoing activity in working memory."

The mechanism behind that borrows from a general model of anxiety, processing efficiency theory, published by Eysenck and Calvo in 1992. In Ashcraft's description, anxiety is thought "to disrupt ongoing working memory processes because anxious individuals devote attention to their intrusive thoughts and worries, rather than the task at hand." For math specifically, he writes, "such thoughts probably involve preoccupation with one's dislike or fear of math, one's low self-confidence, and the like."

Read that again with your own child in mind. The worry is not sitting politely beside the math. It is eating the same resource the math runs on.

The experiment that showed this is worth describing, because it is unusually clean. In 2001 Ashcraft and Kirk gave adults addition problems to do in their heads, and at the same time asked them to hold a set of random letters in memory and recall them afterwards. Two letters on some trials, six on others. The letters were there purely to occupy working memory, so that if the arithmetic started to suffer, you would know the arithmetic had been depending on that same workspace.

It suffered. On the problems that required carrying, errors "increased substantially more for participants with high math anxiety than for those with low anxiety", and the gap widened as the letter task got harder. Give a math-anxious person something else to hold in mind, and their arithmetic degrades faster than anyone else's.

That is what a timed test does. It hands your child a second task, made of worry, and then asks for the carrying.

So the freeze is not a memory failure and it is not a character failure. It is a capacity problem, created in the moment, by the worry.

How to tell anxiety from a gap in knowledge

The mechanism gives you the test, and it costs you one evening.

There is real evidence behind this, and it comes with a limit you need to know about.

Ashcraft and colleagues gave people whole-number arithmetic on paper, untimed, and found no anxiety effects. Other participants got the same problems under time pressure in the lab, and there were "substantial anxiety effects on the same problems." Take the clock away and the gap closes. Two points of honesty about that: these were different groups of participants rather than the same people tested twice, and the timed condition was a lab setup measuring single problems rather than a school test paper.

Now the limit, and it matters for anyone past the early primary years. When Ashcraft's team scored a full achievement test line by line, they found "no math-anxiety effects whatsoever on the first half of the test, which measured performance on whole-number arithmetic problems." On the second half, which brought in "mixed fractions ... percentages, equations with unknowns, and factoring", there was "a strong negative relationship between accuracy and math anxiety." His conclusion is careful: math-anxious people "do not have a global deficit in math competence, and they can perform as well as their peers on whole-number arithmetic problems", but for higher-level material, researchers "do need to take the competence-anxiety relationship into account."

Read that plainly. On basic arithmetic, taking the clock away separates the two things cleanly. On fractions, percentages and algebra, anxiety and real gaps in competence are tangled together, and no kitchen-table check will pull them apart. So the check below is at its most reliable on number facts and arithmetic fluency, and at its least reliable on exactly the topics a Grade 7 or Grade 8 child is likely to be struggling with. Use it as a pointer, and expect a mixed answer.

A timed test applies four things at once: a clock, an audience, real stakes, and no warning about which questions are coming. Take all four away and give the same question again.

What you are collecting here is a set of observations. The right-hand column below describes a pattern that leans towards anxiety. It does not establish it.

That last row is the one parents tend to find most convincing. A gap in knowledge has a shape. It sits on one topic, the one that was taught the week your child was off sick, and it shows up in the same place every time. Anxiety scatters. It takes easy marks near the start of a paper, questions your child can do standing up in the kitchen, and that pattern is hard to explain any other way.

There is a fifth check, and it comes straight from the distinction above. Because math anxiety attaches to the math itself, ask whether your child freezes in every subject's tests or only in this one. A child who goes blank in history and French as well is telling you something broader than a math story, and that is worth raising with the school rather than with a math tutor.

Two cautions. Plenty of children have both at once, and the anxiety is often about the gap, so a mixed result is the normal result rather than a failed test. And if the distress reaches beyond math lessons and math homework, that is a conversation for your family doctor. It is outside what any teacher or tutor should be advising on, and we will leave it there.

"Is this a learning disability?"

Parents ask this early, and it deserves a straight answer rather than a swerve.

Math anxiety and a specific learning difficulty in math are different things. Anxiety is a reaction to the math, and the tests above are what it looks like. A specific learning difficulty is a persistent difficulty with number itself, and it is identified through a formal assessment by a qualified professional, usually arranged through your child's school.

Two things follow from that, and both matter.

Telling them apart is not your job. It is also not a tutor's job, and not a classroom teacher's job. What a teacher can do is describe precisely what they observe: which topics fail, whether the method survives without a clock, whether the errors move around. That description is exactly what a school needs in order to decide whether a formal assessment is warranted. Careful observation is useful. A label from someone unqualified to give one is not.

Ontario names this in the curriculum, then stops reporting on it

Here is something most Ontario parents have never been told, and it changes how you read a report card.

The Ontario curriculum for mathematics, Grades 1 to 8, opens with Strand A, "Social-Emotional Learning (SEL) Skills in Mathematics and the Mathematical Processes". The ministry asks educators to look for chances to teach and reinforce these skills, so that students build confidence, cope with challenges, think critically, and come to see themselves as math learners.

Six skills are named. The second is "recognize sources of stress and cope with challenges."

The elaborations are more specific still. Students should "work through challenging math problems, understanding that their resourcefulness in using various strategies to respond to stress is helping them build personal resilience." They should "recognize that testing out different approaches to problems and learning from mistakes is an important part of the learning process, and is aided by a sense of optimism and hope." They should "see themselves as capable math learners", and take ownership of their own learning.

So the province already knows. This is not a gap in Ontario's thinking.

Now the part that explains your report card. Since the 2021-22 school year, schools have been asked "not to assess, evaluate or report on the overall expectations related to social-emotional learning skills" in that curriculum, while "instruction of the social-emotional learning skills will continue."

The note itself gives no reason, and I could not find published reasoning for it, so I am not going to guess at one. Both halves are true and both are worth holding. The teaching continues. The reporting stopped. Which means a child can spend a year visibly struggling to cope with math under pressure, in a province whose curriculum names that exact skill, and nothing about it will appear on the paper you get sent home. You are not missing a box on the report card. There isn't one.

What the Canadian numbers show, and what they do not

Ontario's own assessment agency asks students how they feel about math, and publishes it alongside the achievement results.

EQAO, 2024-2025 provincial results, English-language schools:

Grade 6 is the floor on both attitude questions. The Grade 3 and Grade 6 figures sit nineteen points apart on liking math and fourteen apart on confidence, and achievement shows a similar gap, 64% against 51% meeting the standard.

Read those gaps carefully. Those are two different groups of children measured in the same year, so the gap is not one cohort declining. It does not follow that any individual child fell nineteen points between Grade 3 and Grade 6. What it does show is that Grade 6 students in Ontario report liking math and feeling capable at math markedly less often than Grade 3 students do, and that the Grade 9 figures sit near the Grade 6 level rather than recovering.

Be careful with what those rows are, though. "I like math" is not a measure of anxiety. A child can dislike math and be perfectly relaxed about it, and an anxious child can like math and dread being tested on it. EQAO asks students about attitude, and attitude is a different measurement. The achievement side of the same reports is covered in what the EQAO levels actually mean.

For anxiety specifically, the international trend runs upwards. Summarising the 2022 PISA round, the GEM Report team reports that the index of mathematics anxiety "increased in a statistically significant way in 37 countries and economies between 2012 and 2022." The same sentence names where the rise was sharpest: Austria, Denmark, Germany, Iceland, Lithuania, Poland, Slovakia, Sweden, Türkiye, the United States and Viet Nam. Canada appears nowhere on that page, in that list or in its list of countries where anxiety fell. So this is not a Canadian finding, and it is not a claim about Canada in either direction.

And there is a hole worth naming. We downloaded Canada's national report on the 2022 PISA results and searched every page of it for six different terms covering anxiety, nervousness and confidence. There is no mathematics anxiety figure in it. The measure exists, Canada participates in the study, and our national highlights report does not carry the number. So when you find yourself unable to look up how common this is among Canadian children, the reason is not that you are searching badly.

The homework finding most parents have never heard

In 2015 a research team led by Maloney studied 438 first and second grade students together with their primary caregivers, and found something that reads uncomfortably at first.

Children of more math-anxious parents learned significantly less math over the school year, but only where those parents helped frequently with homework. The frequency of the help is what carried the effect. That summary comes from the universities' own write-ups of the study, because the paper's abstract page is behind a cookie wall.

The control in that study is what makes it convincing. Reading achievement showed no relationship to parental math anxiety at all. So this is not a general story about stressed households or about anxious parents raising anxious children across the board. It is specific to math, and it travels through the homework table.

Please do not read that as an instruction to stop helping. The study does not say that, and withdrawing from your child's homework is its own kind of message. What the finding points at is how the help happens. If math at the kitchen table reliably ends with a raised voice, a sigh, or "I was never any good at this either", the frequency of that help is doing work you did not intend. Changing the tone is one option. Having someone else do the explaining, so that your role at the table stops being the person who knows the answer, is another.

What a teacher does about it

A teacher working with a child who freezes separates the math problem from the pressure problem, because those need different work. They rebuild the specific topic where the marks actually cluster, which is a knowledge job. Then, separately, they practise the method under conditions that gradually start to resemble a test, so that the retrieval itself gets rehearsed rather than only the content.

They also watch for avoidance, which is the expensive part. Math compounds. A child who quietly stops attempting a topic in Grade 6 meets it again in Grade 8, and Grade 9 math in Ontario is now a single de-streamed course that everyone takes, so there is no easier lane to slip into. What that transition assumes your child already has is the thing avoidance quietly removes.

The narrowing comes later, and it is sharper than most families expect: the Grade 11 course is the real decision, because it governs which Grade 12 math courses your child is eligible for at all. Avoidance in Grade 6 is how a bad term becomes a closed door five years later, which is why interrupting it early is so much cheaper.

And they make the format boring by rehearsing it, so that sitting the real thing is familiar rather than novel. If the Grade 9 assessment is ahead of you, our twenty practice questions with worked answers are a reasonable starting point. They are original questions rather than reprinted ones, and the page deliberately tells you to work through them on paper first with no timer, which is the right order: get the method solid, then add the clock.

None of this treats anxiety, and none of it promises a mark. It is teaching, arranged around what the research says the pressure is doing to a child's working memory.

If the same pattern shows up in your child's reading, the shape of the problem is similar and we have written about it separately.

Where Genie fits

Genie works only with Canadian Certified Teachers, licensed by a provincial governing body such as the Ontario College of Teachers. Certification is verified before a teacher ever sees a student profile: the certificate number and issuing body are checked against the provincial registry, with a live government-ID match. We call that the Trust Engine.

For a child who freezes, two parts of how this works are worth knowing. Teachers are matched by course, pace and personality, and for an anxious child those last two deserve more thought than they usually get. The research here is suggestive rather than settled. Ashcraft points to a study by Turner and colleagues which documented student avoidance under teachers who demanded correctness while giving little support, one of whom "often showed annoyance when students gave wrong answers", and those researchers speculated that such students may feel "vulnerable to public displays of incompetence". Ashcraft's own summary is that nobody has yet done thorough work on where math anxiety comes from. So how an adult reacts to a wrong answer is a reasonable thing to care about, on suggestive evidence, and not a proven lever.

The second part is the per-lesson record. Parents get a record of what was covered and what the child understood. If your child freezes, you already know how those after-school conversations go, because "how was it?" gets you "fine" and nothing else. A written record means you can follow whether the specific topic is moving without interrogating a tired ten-year-old about their feelings.

One more thing worth knowing, given where avoidance leads. A certified teacher does not only work through the coursework in front of them. They can chart which courses a student needs in order to reach a particular programme later, and then aim the teaching at the specific areas that get them there. For a child who has been quietly avoiding a topic since Grade 6, that pairing matters more than it sounds: somebody is tracking both the gap and the door it eventually closes.

You can find a certified math teacher here.