Parents Who Help With Maths Homework Can Make Maths Anxiety Worse. The Research, and the Fix.
If you are anxious about maths and you help with maths homework often, your child learns less maths over the school year and becomes more anxious about it - more so than children of equally anxious parents who help rarely.
That’s the finding from a 2015 study in Psychological Science. It is not an argument for abandoning your child to their worksheet. It’s a specific result with a specific mechanism, and once you see the mechanism the fix is fairly obvious.
What the study found
Maloney, Ramirez, Gunderson, Levine and Beilock followed first- and second-graders across a school year, measuring the children’s maths achievement and maths anxiety, and separately measuring their parents’ maths anxiety and how often those parents helped with maths homework.
The headline: children of maths-anxious parents learned less maths over the year and were more likely to become maths-anxious themselves.
The crucial qualifier: only when those parents helped frequently with homework. Where maths-anxious parents helped rarely, their anxiety showed no relationship with their children’s learning or their children’s anxiety at all.
It was the first study to establish that maths anxiety passes from parent to child - and to identify homework help as the channel it travels through.
The detail that matters: it’s the frequency, not the parent
Read that qualifier again, because almost every summary of this study drops it and the study is close to meaningless without it.
Being anxious about maths does not harm your child. Plenty of maths-anxious parents in that sample had children who were fine. What predicted the damage was anxiety plus regular, hands-on involvement in homework.
Which tells you the mechanism isn’t genetic and isn’t about ability. It’s transmission through contact. Sitting down nightly with a subject that makes you tense, you communicate a great deal you never say aloud - the sigh before you look at the page, “I was never any good at this either”, the visible relief when it’s over, the impatience that is really discomfort. A seven-year-old reads all of it accurately.
They don’t learn that maths is hard. They learn that maths is the thing that makes the adults in this house unhappy, and that is a far stickier lesson than anything on the worksheet.
The other finding, which cuts the opposite way
There’s a second body of work worth knowing: on average, parental help with homework in the primary years shows no measurable benefit to achievement at all.
So the picture is less “help is good, anxious help is bad” and more: the homework hour is doing considerably less than families think it is, and for a meaningful minority it is doing active harm. That’s a lot of evenings for a result that rounds to zero.
What this does not mean
Three things it is not saying, because the finding gets over-read in both directions.
It is not saying be uninvolved. Parental involvement in a child’s education broadly is well-evidenced and valuable. What’s in question is one specific activity: sitting beside them doing the maths.
It is not saying your anxiety is a failing. Maths anxiety is extremely common, it usually traces back to someone else’s classroom twenty-five years ago, and it is not something you can decide your way out of before Thursday’s homework.
It is not a licence to disengage. “The research says I shouldn’t help” is a convenient misreading. The research says this particular form of help, delivered with this particular feeling, backfires - and it points at what to do instead.
What to do instead: be a bad audience, on purpose
The useful move is to change what you are in the room, not whether you’re there. Stop being the person who knows, and become the person it gets explained to.
That’s not a consolation prize. Explaining is one of the most reliable ways to learn anything - in cross-age tutoring studies the person doing the teaching gains slightly more than the person being taught - and it requires an audience who genuinely doesn’t know. You are extremely well qualified for that role.
In practice:
Don’t know the answer, visibly. “I genuinely don’t remember how this works - walk me through it” is a better opening than “show me what you did”. The first asks for an explanation. The second asks for a performance, with you as examiner.
Don’t correct. Ask. When something’s wrong, “hang on, I didn’t follow that bit” does the work that “no, it’s actually…” undoes. You’re pointing at the gap rather than filling it. This is hard and you will fail at it regularly.
Ask for the wrong version. “What mistake might someone make here?” A child can produce a right answer they don’t understand; they can’t construct a plausible error without knowing where the reasoning forks. It’s the most honest check available at a kitchen table.
Let the silence run. The pause where they realise they can’t finish the sentence is the entire point. Rescuing them from it is the commonest way a well-meaning parent dissolves the benefit.
Ten minutes, not an hour. This is effortful in a way re-reading isn’t - which is why it works, and why it doesn’t scale to a long session.
And notice your own face. You don’t have to perform enthusiasm. Neutral is plenty. What transmits is the sigh, not the sentence.
If you aren’t maths-anxious
Then most of this doesn’t apply to you, and you can help freely - though the “no measurable benefit” finding still suggests the hour is worth less than it feels.
Even so, the bad-audience approach is worth adopting on its own merits. A parent who explains well produces a child who has watched someone explain well. A parent who listens badly on purpose produces a child who has had to explain it themselves, which is the thing that actually sticks.
When to hand it over
If maths homework has become the worst half-hour of the day in your house, the most effective intervention may simply be that the help stops coming from you.
Not because you’re doing it wrong - because the relationship is carrying something it shouldn’t have to. A child who is embarrassed to say “I still don’t get it” to a parent will often say it immediately to someone whose approval isn’t at stake. That’s a large part of what one-to-one tutoring is buying, and it’s worth being honest that it’s a reason.
Frequently asked questions
Should I stop helping with maths homework?
Not necessarily. If maths doesn’t make you anxious, help as you like - though the evidence suggests it does less for achievement than you’d expect. If maths does make you anxious and you’re helping nightly, that specific combination is the one associated with worse outcomes.
What did the 2015 study actually show?
Maloney and colleagues found that children of maths-anxious parents learned less maths across the school year and grew more maths-anxious - but only where those parents helped frequently with homework. Where help was infrequent, parental anxiety showed no relationship with either outcome.
Does parental homework help work at all?
On average, in the primary years, it shows no measurable effect on achievement. That doesn’t make involvement pointless; it makes this one activity less powerful than families assume.
I am anxious about maths. Have I already damaged my child?
Almost certainly not. The anxiety on its own wasn’t the predictor - anxiety combined with frequent hands-on help was. And the mechanism is behaviour, which is changeable in a way that temperament isn’t.
What should I say instead of explaining?
“I don’t remember how this works - walk me through it.” Then: “I didn’t follow that bit.” Then: “What mistake might someone make here?” Three sentences, and they do most of the work.
My child is in secondary school. Does this still apply?
The study looked at younger children, so treat it as directional for teenagers. The underlying mechanism - anxiety transmitting through contact, and explanation beating being explained to - isn’t age-specific.
Related reading
- The Protégé Effect: Why Children Learn More From Teaching Than From Being Taught - the research behind being a bad audience on purpose.
- Understands It in Class, Blanks in the Exam - how to tell nerves apart from the three problems that look like them.