Problem-Solving Skills for Kids: You Can’t Teach Them. You Can Only Stop Preventing Them.

The reason your kid can’t solve a problem on her own is that somebody has always solved it first, and I say that with love, because for eleven years the somebody was me.

I taught high school English, and I can tell you exactly what happened every time I gave an assignment with no rubric and no model. They stared into space. Then the hands went up, all at once: What do I do? Where are the directions? What should I write about? How can I do this if you haven’t given me a model? These were not weak students. Plenty of them were honors kids. But every problem they had ever been handed came with the method attached, and a problem with the method attached is not a problem. It’s an instruction. By fifteen they had done thousands of instructions and almost no problems, and an open page looked to them like a mistake I had made.

What problem-solving skills for kids actually are

If you searched this looking for a list of skills to teach, problem-solving isn’t a skill you teach, in the sense of explaining it and having a child practice the explanation. It’s a capacity a child builds by being left alone with problems that are hers, at a size she can handle, often enough that she learns she can. The pieces people put on the lists (identify the problem, brainstorm options, try one, check whether it worked) are what a child does naturally when nobody rushes in. The lists describe the behavior of a kid who has been allowed to struggle. They don’t produce one.

What produces one is struggle that’s allowed to run. A child who wants the toy on the high shelf and is left to figure it out will drag over a chair, fail, get a bigger chair, fail, stack a box on the chair, and get the toy, and in the process she has done every step on every list, and nobody taught her anything. A child whose parent hands her the toy at the first whine has done none of them, and will whine faster next time, because whining worked.

Why struggle first works (the research is clear on this one)

This is one of the better-studied ideas in learning science, and the finding runs against every instinct a loving adult has. Manu Kapur, a learning scientist now at ETH Zurich, has spent nearly two decades on what he calls productive failure: give students a hard, open problem with no help first, let them flail and produce messy, wrong solutions, and only then teach them the method. Students who go through that order consistently understand the concept more deeply, and transfer it to new problems better, than students who are taught the method first and then practice it. The flailing isn’t wasted. It’s the part where the mind builds the hooks that the later explanation hangs on.

Kapur’s work is with older students and math, but the same pattern shows up at six. Jane Barker’s team at the University of Colorado found in 2014 that the more of a young child’s week was unstructured, without an adult setting up and running the activity, the better she was at self-directed executive function, which is the everyday name for solving your own problems. The kids with the most adult-run time were the worst at it. Every adult-run hour is an hour in which the adult did the problem-solving.

Read those two together: children learn to solve problems by having problems and not being rescued from them, and the rescuing is the thing we‘re best at.

What I watched it become

The students with their hands in the air had simply never, in fifteen years, been handed a problem that was actually theirs to solve, and so the muscle wasn’t there. They experienced an open question as anxiety. They wanted to know what I wanted before they’d think, because in their experience thinking was a way of getting it wrong, and getting it wrong was a catastrophe rather than a step. The ones who surprised me, year after year, were rarely the top of the grade book. They were the kids who struggled and kept going anyway, my English language learners especially, who were working twice as hard for every sentence and did not quit. Persistence in the face of a hard problem turned out to have almost nothing to do with the transcript.

That’s the systems layer. A school day organized around one right answer, delivered by an adult and checked against a key, is a day with no room for productive failure in it, because failure isn’t productive when it’s graded. And a childhood organized around adult-run activities has the same shape at home. The kid who arrives in ninth grade unable to start is the product of both. Blame the design, not the kid, and not yourself for having absorbed it.

What this looks like in a house

In my house I don’t step in to help, and so my kids teach themselves things constantly. My oldest daughter recently decided she wanted banana muffins, taught herself to make them from scratch, and then took the job on for the family; she makes them every week now, no help, no recipe card from me. Nobody assigned that. She had a problem (she wanted muffins) and I didn’t solve it for her.

In our house the rule when you’re bored is that you know the options and a screen isn’t one of them, and that rule turns out to be a problem-solving engine, because an empty afternoon is a problem and a child left alone with it solves it. So is a mud kitchen with no recipe. So is a tree with no ladder: my kids climb high trees, roll down hills, jump off rocks and balance along branches as a matter of routine, and every one of those is a problem-solving session (how do I get up, how do I get down, is this branch going to hold me) with no adult supplying the answer. None of it is a lesson. It’s what happens when you don’t prevent it.

How to teach problem solving, if you insist on a method

Stop solving. That’s most of it, and it’s the hard part. Then, when she brings you the problem, hand it back with a question instead of an answer: “What have you tried?” “What else could you try?” “What do you think would happen if…?” Then be quiet. The silence is where the thinking happens, and most of us fill it within three seconds.

Let it be wrong. A tower that falls is a better lesson than a tower you steadied. A lost game she planned badly teaches more than one you fixed. Kapur’s students learned from producing wrong answers first, and yours will too, as long as wrong isn’t treated as a disaster.

Give real problems, not manufactured ones. Not worksheets about problem-solving but a lunch to make, a broken thing to take apart, a disagreement with a sibling to settle without you, a route to plan. Real problems have real stakes and real feedback, and a child can tell the difference.

And size them. A five-year-old’s problem is the zipper; a ten-year-old’s is the week’s schedule for her own work. Too big and she gives up, too small and there’s nothing to solve. If you homeschool, the day is full of problems that can be hers, and handing her ownership of her own list is the whole idea behind the Planner and the Student Dashboard that comes with it.

One thing

The next time your kid brings you a problem, count to ten before you say anything, and then ask what she’s tried. Notice how often the answer arrives before you get to ten.

I write every week about what childhood is for and how to protect it. You can find me on Instagram or join the list.

Sources

• Kapur M, “Productive Failure,” Cognition and Instruction 26(3), 2008. Link
• Sinha T, Kapur M, “When Problem Solving Followed by Instruction Works: Evidence for Productive Failure,” Review of Educational Research 91(5), 2021. Link
• Barker JE, Semenov AD, Michaelson L, Provan LS, Snyder HR, Munakata Y, “Less-structured time in children’s daily lives predicts self-directed executive functioning,” Frontiers in Psychology 5:593, 2014. Link