Should you help with your kid's math homework?

The 30-second version.

The advice-column answer — “step back, let them struggle” — is too blunt to be useful, and it leaves a lot of involved, well-meaning parents feeling vaguely guilty for sitting at the table. The research is more specific and, honestly, more freeing. The best-known finding (Maloney et al., 2015) isn't that helping is bad. It's that a math-anxious parent who helps frequently tends to transmit the anxiety — and a calm one who helps does no such harm. Pair that with the broad review of parental involvement (Pomerantz et al., 2007), whose subtitle is literally More Is Not Always Better, and it resolves: presence isn't the variable. The kindof presence is. Be the calm one who helps with the process, not the answers — or, if math winds you up, change the mode of help. That's the whole post.

The finding everyone half-remembers.

The study people are vaguely gesturing at when they say “parents pass on math anxiety” is Maloney et al. (2015), in Psychological Science. It followed 438 first- and second-graders and their parents across a school year. The headline is real, but it's narrower and more interesting than the version that circulates.

What they found: when parents were more math-anxious, their kids learned significantly less math over the year and ended it more anxious about math themselves — but only when those anxious parents reported helping with homework frequently.Among math-anxious parents who helped less often, there was no such relationship. The effect was specific to math (it didn't show up for reading), and the broader link between a parent's math anxiety and a child's math outcomes has held up in later work.

Read the mechanism carefully, because it's the hinge of the whole question. The problem the study identifies is not“a parent helped.” It's “an anxious helper transmitted the anxiety.” The frequent help is the channelthe anxiety travels through — the nightly dose of a parent's dread and “ugh, I was never good at this either” — not the cause of the harm by itself. (We unpack the cognitive side of that anxiety in the math-anxiety piece.) So the conclusion to draw is the opposite of “don't help”: mind what you're carrying into the room when you do.

So the question isn't whether. It's how.

This is where the second body of research earns its place. Eva Pomerantz and colleagues reviewed decades of work on parental involvement in children's academic lives (Pomerantz, Moorman & Litwack, 2007), and the title says the finding out loud: More Is Not Always Better. Involvement helps or hurts depending on its form. The autonomy-supportive kind — following the kid's lead, focusing on process, keeping the affect warm — tends to help. The controlling, intrusive, pressure-laden kind tends not to, and can backfire.

Stack the two studies and they point the same direction. Maloney says the danger is the emotional content a helper brings; Pomerantz says the payoff depends on the styleof help. Neither says “leave.” So you don't have to choose between hovering over every problem and abandoning them to it. The actual move is to stay — and change what you're doing while you're there.

What good help looks like.

The throughline: help with the relationship and the process, not with the answers. You are a thinking partner, not an answer key. Concretely, that's a handful of moves.

Be the calm presence.The most valuable thing you provide is often not mathematical at all — it's a regulated adult in the chair next to a frustrated kid, the hard thing being done alongside someone safe. That's help, and it's the kind the research likes.

Ask “walk me through it” instead of supplying the answer.A question keeps the kid in the driver's seat and shows you the actual reasoning, so you can find the real slip together instead of papering over it. (There's a whole piece on what to say in the moment your kid is wrong; the short version is: pause, ask, don't correct at the speed of reflex.)

Normalize the struggle.“This part is genuinely hard” is more honest and more useful than “this is easy, you've got it.” A kid who's told an obviously hard thing is easy just concludes the problem is them.

Stop before the frustration peaks.The instinct to push through “just these last three” while a kid is melting down is almost always wrong. Ending the session a few problems early, on a calm note, protects the relationship with math — which is the asset you're actually guarding. The break is not giving up; it's good pedagogy.

Don't take over the pencil.The literal moment your hand reaches for the pencil to “just show them” is the moment the learning leaves the kid. Narrate, point, ask — but let them make the marks. Their work, their hands, their thinking.

What backfires.

The mirror image, drawn from the same two findings. These are the forms of “help” that the research suggests cost more than they give.

• Doing it for them. Finishing the problem, supplying the number, walking them line by line to the answer. It feels efficient and kind; it removes the part that was the point. The kid learns that waiting produces the answer.
• Speed pressure.“Come on, this should be quick.” Speed isn't understanding, and pushing for it eats the working memory the kid needs for the math — the exact mechanism under math anxiety.
• “Ugh, I was never good at math either.”Meant as solidarity, it lands as inheritance — you've just modeled the identity and handed it down. (This is the precise sentence the “I'm bad at math” piece flags as a transmitting one.)
• Turning help into a power struggle. When the homework becomes a battle of wills — comply, sit up, do it my way — the math is no longer the subject; control is. Nobody learns fractions inside a standoff.
• Hovering that's really anxiety-driven.The over-correcting, over-checking, can't-leave-the-table kind of help is frequently the parent's anxiety wearing a helpful costume — the controlling involvement Pomerantz warns about and the frequent-anxious-help channel Maloney warns about, in one move.

What if you're math-anxious yourself?

A lot of parents are — math is the one subject adults openly brag about being bad at — so this deserves a straight answer rather than a guilt trip. The Maloney finding can read like a verdict (“anxious parents harm their kids”), but the actual implication is a fork, and both paths are fine.

Option one: split the roles.Outsource the procedural, this-is-how-you-borrow-across-a-zero help to someone who isn't bringing your dread to the table — a partner, an older sibling, a tutor, a tool, the teacher during office hours — and keep, for yourself, the role you're uniquely good at: the emotional-safety person, the calm one who believes in the kid regardless of tonight's worksheet. That isn't opting out of parenting. In the Maloney framing it's removing the transmission channel while keeping everything good about your presence.

Option two: help, but regulate consciously.If you want to be the one in the chair, you can be — the study's harm runs through the anxiety, not the help. So don't narrate your own dread: don't sigh, don't editorialize about how you always hated this, don't telegraph that the problem is scary. A plain, curious “huh, I don't remember how this one works either — let's figure it out together” is a gift. It models curiosity in the face of not-knowing, which is the actual skill, instead of modeling avoidance. Kids learn more from watching you be calm about being stuck than from watching you already know the answer.

If your kid is neurodivergent.

One honest note, because the “back off and let them struggle” advice can mislead badly here. For a kid with ADHD or dyscalculia, productive homework help is often more about structure and co-regulation than about math content: breaking the work into small pieces, sitting alongside as a steady presence (body-doubling), holding the pace and the transitions, keeping the affect calm when frustration spikes. Pulling that scaffolding away in the name of independence can leave a kid stranded.

And the parent's calm matters even more for these kids, not less. A child who already finds the math genuinely harder is reading your face for whether this is a safe place to be stuck — and the most useful help you can give is often to make sure the answer they read there is yes.

One last thing.

So — should you help with your kid's math homework? Yes, if you can be the calm one; in a different mode, or through someone else, if math winds you up. Either way, the thing you're really protecting isn't tonight's right answers but your kid's relationship with a hard subject — and with you while they do it. The number on the page matters less than whether the kid walks away from the table still willing to try the next one, and there's no shame in whichever way you find to make that true.

If you want to hear when we've got more for parents, drop your email at the waitlist. We email a few times. Total. From a real person.