STEM for Little Kids: What Actually Builds a Young Thinker

Search "STEM for toddlers" and you'll find coding robots, science kits and apps promising to turn a three-year-old into an engineer. The research on how young children learn points somewhere less expensive and far more ordinary: a child who asks a lot of questions, gets real answers, builds things with their hands, and is asked to explain what they think.

This article covers what the evidence says, what it doesn't say, and what you can do this week at home.

What "STEM" means at age 3

For a preschooler, STEM is not a subject. It is a handful of habits:

• Noticing and wondering. "Why is the moon following our car?"
• Guessing before knowing. "I think the big block will fall first."
• Trying it. Stacking, pouring, mixing, dropping, sorting.
• Explaining. "It fell because the bottom one was too small."
• Early maths. Counting, comparing ("which has more?"), shapes, patterns, and where things are ("under", "behind", "next to").

None of these needs a kit. All of them need an adult who is paying attention.

Early maths matters more than most parents think

The best-known study here is a 2007 analysis led by economist Greg Duncan, which combined six large long-term studies of children in the US, UK and Canada. Children's skills were measured when they started school, around age 5, and their achievement again between ages 7 and 14.

Of everything measured at school entry, early maths skills, such as knowing numbers and understanding their order, were the strongest predictor of later success. Early reading came next, then attention skills. One finding surprised the researchers: early maths predicted later reading better than early reading predicted later maths.

Two cautions:

• "Predicts" is not "causes". The study shows a link, not proof that drilling numbers into a four-year-old will change their future.
• It is not a case for worksheets. The skills that matter here are number sense, comparing quantities and understanding order. Children build these through play, conversation and everyday counting far more than through flashcards.

Blocks are serious business

If there is one "STEM toy" with good research behind it, it's a box of blocks.

In a study of more than a hundred 3-year-olds, children who were better at copying block structures were also better at early maths. The skills involved are working out whether a block goes above or below another, and whether the pieces line up. The researchers also found that by age 3, children from lower-income families were already behind on these spatial skills, most likely because they had had less experience with blocks and similar toys.

Other studies connect preschool block play with later maths achievement, and a more recent trial with preschoolers from low-income families found block play promising for geometry, spatial skills and self-regulation. As with the maths research, much of this is correlational, so treat blocks as a very good bet, not a guarantee.

What helps most is play with a little structure. Build something and ask your child to copy it. Give them a challenge ("Can you build a bridge a car can go under?"). And use spatial words as you play: on top, beside, between, taller, wider, half.

Your child's questions are the curriculum

Young children are relentless question-askers. In one well-known study, preschoolers talking with adults asked an average of 107 questions an hour. Most of these are fact questions ("What's that?"), and fewer are "how" and "why" questions.

These questions are not noise. Researchers have found that they are genuine attempts to fill gaps in what a child knows, and that children use the answers they get. When a child receives a vague non-answer, they tend to ask the same question again. When they get a real explanation, they move on to a new, deeper question.

So the most useful STEM habit at home costs nothing: take the question seriously. Give a short, true answer. When you don't know, say "Let's find out." Sometimes turn it around: "What do you think?"

Asking your child to explain makes them think better

This is the most useful finding in this article, and the least well known.

In a series of experiments, children aged 3 to 6 were shown a new toy, such as a machine with interconnected gears. Some were asked to explain how it worked, and others simply described what they saw. The children who explained learned more about how the toy worked, such as which gears drove which, but not more about surface details such as colour. In another study, 5-year-olds who were asked to explain their observations were better at picking the idea that fit more of the evidence.

A review of this research concludes that generating explanations deepens children's understanding of cause and effect and helps them apply what they learned to new situations. It works especially well when something unexpected happens. Explaining also shows a child where the gaps in their own knowledge are.

At home, this means one small question after almost anything:

• "Why do you think that happened?"
• "How did you know?"
• "What would happen if we tried it again with the big one?"

Don't correct the answer straight away. The act of explaining does the work, even when the explanation is wrong.

"Serve and return": the engine underneath all of it

The Harvard Center on the Developing Child uses the phrase serve and return for the back-and-forth that builds young brains. A child "serves" with a word, a point, a question or a look, and an adult "returns" it by noticing, responding and naming what the child is focused on. The Center describes these exchanges as helping to build the brain connections behind communication and social skills.

Every idea in this article is a form of serve and return. A question answered, a block tower built together, a "why do you think?": each one is a child serving and an adult returning.

Where screens fit, honestly

Young children learn less from a screen than from a person. Researchers call this the "video deficit". A meta-analysis of 59 studies of children aged 0 to 6 found that children learned noticeably less from video than from the same thing shown in person. The gap shrinks as children get older.

What the paediatricians say. The American Academy of Pediatrics' long-standing guidance is no screens except video chat before about 18 months, and roughly one hour a day of high-quality content, ideally watched together, for ages 2 to 5. In 2026 the AAP shifted its emphasis from strict time limits towards quality, context and conversation. The main concern is what screens push out of the day: sleep, active play, and time with family.

"Educational" on an app store means very little. A 2021 study examined the 100 most-downloaded children's "educational" apps on Google Play and the Apple App Store, plus 24 apps that preschoolers actually used. Scores were low across the board, and free apps scored lower still, largely because of distracting extras such as ads and pop-ups. The researchers' advice to parents was not to assume an app labelled educational actually is.

What a good app looks like. The researchers used a framework known as the "four pillars": a good learning experience is active (the child thinks, not just taps), engaged (no distractions pulling them away), meaningful (it connects to their real life) and socially interactive (it involves other people, or leads to a conversation with you). Use those four as a checklist for any app your child uses.

The takeaway is not "screens bad". A screen should be a small, well-chosen part of the day, and it works best as a starting point for a conversation, not a replacement for one.

A week of real STEM at home

No purchases needed.

Notice the pattern: wonder, guess, try, then say why. That loop is the core of scientific thinking, and a three-year-old can do all four steps.

The short version

1. Early maths (counting, comparing, patterns, space) matters, and play teaches it best.
2. Blocks and puzzles are among the best-supported "STEM toys" there are.
3. Treat your child's questions as the lesson, and give real answers.
4. Ask "why do you think that?" often. Explaining builds understanding.
5. Keep screens small, choose them carefully, and talk about what's on them.

You don't need a robot kit. You need blocks, a bucket of water, and time to listen.

This article is general information for parents, not medical or educational advice for a specific child. If you have concerns about your child's development, talk to your paediatrician.