How to help with math homework without a fight

If math homework regularly ends in tears, the most useful change is usually to help less often and differently โ not to try harder. Short answer: a study of 438 first- and second-graders found that children of math-anxious parents learned significantly less math across the school year, but only when those parents helped with homework frequently. When they helped rarely, the effect disappeared. Your job at the table is to keep the evening short and calm, not to make every answer right.
Why can helping with math homework backfire?
Researchers at the University of Chicago followed 438 first- and second-graders and their parents across a full school year (Maloney and colleagues, Psychological Science, 2015). Children whose parents reported high math anxiety learned more than a third of a grade level less math than their classmates over that year, and finished it more anxious about math themselves.
Here is the part that matters, and it is the hopeful part: that gap only appeared where the math-anxious parents said they helped with math homework often. Where they helped rarely, their own anxiety had no measurable link to how their children did. The anxiety was not reaching the child through the air. It was reaching them across the homework page.
That is worth sitting with for a second, because it quietly rewrites the job description. The problem is not that you are bad at math. The problem is a particular kind of help โ frequent, corrective, and tense โ delivered by someone the child trusts completely.
How would I know if I'm passing on my own math anxiety?
Almost nobody does this deliberately, and it rarely looks like anger. It usually sounds like small, throwaway remarks:
- "I was never any good at math either." Said kindly, meant as comfort, heard as permission to stop trying.
- "That's not how you do it โ let me show you the proper way," when the child's method was slower but perfectly correct.
- Reaching for the pencil. Even once. It changes whose page it is.
- Sighing, checking the clock, asking "how many left?" every few minutes.
- Visible relief when it's over. Children read that accurately: so that was as bad as I thought.
None of these is a character flaw. They are what tension does to a person who wants the evening to be over. Noticing them is most of the fix.
What should I do instead of correcting?
Four changes, in the order they are worth trying.
- Don't take the pencil. If you write even one digit, the page stops being your child's. Point at the row and wait instead.
- Ask "how did you get that?" before you say anything is wrong. Sometimes the answer is right by a route you did not expect. When it genuinely is wrong, children usually catch the slip themselves halfway through explaining it โ which is worth far more than being told.
- Decide when it ends before it starts. "We'll work until the timer runs out" is a promise you can keep. "We'll work until it's finished" is a promise the evening gets to decide.
- Separate practicing from checking. Practicing is your child's job and it is allowed to be messy. Checking is the teacher's job. Trying to do both at the same table at the same time is what turns ten minutes into an hour.

How long should math homework actually take?
In the US the usual benchmark is the 10-minute rule: roughly ten minutes of homework per night per grade level, across all subjects combined โ not per subject. It comes out of Harris Cooper's research at Duke University and has been endorsed by both the National Education Association and the National PTA since the mid-1990s. Plenty of districts have written it into policy.
| Grade | All homework, per night | If it's regularly longer |
|---|---|---|
| Kindergarten | No set homework; shared reading instead | Ask the teacher what the task is meant to practice |
| 1st grade | About 10 minutes, all subjects | Stop at the limit and note it on the page |
| 2nd grade | About 20 minutes, all subjects | Two rough nights in a row is worth a message to the teacher |
| 3rd grade | About 30 minutes, all subjects | Ask whether a shorter set would still meet the goal |
The exact number is not the point. The ending is. A child who knows the evening has a floor and a ceiling starts faster and argues less, because there is nothing to negotiate.

What if it still ends in tears?
Stop. Not as a punishment, not as a reward โ just stop. A crying seven-year-old is not learning arithmetic, and a page finished in tears teaches something about math that will outlast the page by years.
Then write one honest line for the teacher: "We stopped after 15 minutes โ this was too hard tonight." That is useful information, not a confession. Homework at this age is largely feedback for the teacher about where the class actually is.
If most nights end this way for more than a few weeks, raise it โ with the teacher first. Persistent distress around numbers is a different thing from a hard week, and schools have options. We wrote about telling those apart in struggling with math or dyscalculia, and about the child's side of the same problem in math anxiety in young children.
Where our app fits, honestly
There is a second study worth knowing about, from the same University of Chicago group, published in Science the same year. Families were given a short nightly math activity to do together on a tablet โ a few minutes, a small story with a number question in it. Children's math achievement improved over the school year, and the gain was largest for the children whose parents were the most math-anxious.
The interesting part is not the tablet. It is that the format quietly did three things a worried parent finds hard to do alone: it kept the math short, it gave the session an obvious ending, and it moved the adult out of the role of examiner. That is the honest case for an app like ours at this age. Not that it teaches better than you do โ it doesn't โ but that it lets you sit beside your child during math without being the person marking it.
It will not close a real gap on its own, and it is not a substitute for the teacher. If you are weighing options, we wrote about what actually matters in how to choose a math app for your kid.

Questions parents ask
Should I stop helping with math homework altogether?
No. The research points to helping less often and differently, not to withdrawing. In the study, children whose parents helped rarely showed no link between the parent's math anxiety and their own learning โ but that is not the same as no support at all. Stay in the room, keep the session short, and let your child hold the pencil.
My child gets the right answer a strange way. Should I correct the method?
Usually not on the spot. Ask how they got there first. A slower method that works is a real method, and children often abandon a perfectly good approach simply because an adult looked worried. If the school is teaching one specific method, raise it as a question for the teacher rather than as a correction at the table.
How long should first-grade math homework take?
In the US the common benchmark is about 10 minutes per night per grade level across all subjects โ so roughly 10 minutes in 1st grade, 20 in 2nd and 30 in 3rd. If your child regularly needs far longer than that, it is information the teacher wants, not something to push through.
I'm anxious about math myself. Am I bound to pass it on?
No. In the same research the link only showed up where math-anxious parents helped frequently, and a follow-up study found that a short, structured nightly math activity helped exactly those children the most. What you say about math in front of your child matters more than how good you are at it.



