Math Games for Kids: What the Research Actually Says About Play-Based Learning
A cheap board game and a two-week study that changed how researchers thought about early maths
In 2007, researchers at Carnegie Mellon University ran a study that’s still cited constantly in early maths research today, and it started with something almost embarrassingly simple, a homemade board game that looked a lot like the bottom row of Chutes and Ladders.
Geetha Ramani and Robert Siegler recruited preschoolers from Head Start centres, serving families from lower-income backgrounds, and split them into two groups.[1] One group played a linear board game with ten numbered squares in a row, moving a token forward while counting each square out loud. The other group played what looked like an identical game, except the squares were coloured instead of numbered, so there were no numbers involved at all, just colour matching.
Each child played for around four short sessions, fifteen to twenty minutes each, over two weeks. That’s it. No curriculum, no workbook, no formal lesson.
The results were striking. Children who played the numbered board game made significant gains across four different measures of number knowledge, counting accuracy, numeral identification, comparing which of two numbers was bigger, and estimating where a number belonged on a line. The children who played the colour version showed no improvement on any of these measures at all.[1]
Even more notably, when researchers checked back in nine weeks later, the gains from the numbered game had held. This wasn’t a short-lived novelty effect, it was a genuine shift in how these children understood numbers, built from a few short sessions of play.[1]
That study is worth telling in full because it’s one of the clearest pieces of evidence in early childhood research that a specific, simple game, played briefly and repeatedly, can move the needle on real mathematical understanding. It’s also become a touchstone for a much bigger conversation about what actually makes a math game work, and what doesn’t.
Not every game “counts” as a math game
The board game study is often summarised as “board games help kids learn maths,” but that’s a bit too generous to games in general. The colour version of the exact same board, played for the exact same amount of time by children the exact same age, did nothing measurable at all.[1]
That distinction matters more than it might seem. Researchers who study play-based mathematics have argued that for a game to genuinely build maths skills, a few conditions need to be true. The mathematical content needs to actually be built into how the game works, not bolted on as an afterthought. It needs to be mathematically accurate. It needs to connect to skills the child will keep using later. And it needs to sit at roughly the right level of challenge for that particular child, not so easy it’s mindless, not so hard it’s frustrating.[6]
In plain terms, a maths game works when the maths is the game, not a task bolted onto the side of it. Counting squares while moving a token is the game. Matching colours while incidentally being near some numbers is not.
The evidence isn’t limited to one study
The Head Start board game study is famous, but it’s not alone. A broader systematic review and meta-analysis of game-based learning in early childhood education, published in 2024, pulled together the existing research and found that game-based learning has a moderate to large positive effect across cognitive, social, emotional, and motivational outcomes in young children.[5]
Longer, curriculum-level programs show similar results. The Building Blocks preschool mathematics curriculum, developed by researchers Douglas Clements and Julie Sarama, weaves structured play, small group activities, and simple computer games into everyday preschool teaching. In a controlled evaluation, children using the Building Blocks approach made significantly greater progress in numeracy, geometry, measurement, and pattern recognition than children in a comparison group.[3]
Not every result in this field is a clean win, though, and it’s worth being honest about that. A later study tried to take the success of the numbered board game out of the lab and into actual homes, sending Head Start families home with simple maths card games to play together. The number comparison game, which had worked well in a classroom setting, showed no measurable improvement in this home version. Interestingly, a separate shape and colour matching game sent home in the same study did improve children’s shape knowledge.[2] The researchers noted wide variation in how parents guided their children through the games, which may explain some of the difference.
That’s a useful, if slightly humbling, finding for anyone building a shop full of printable and play-based resources. A game that works beautifully in a research setting doesn’t automatically translate to a busy kitchen table, and how an adult plays alongside a child seems to matter just as much as the game itself.
The bigger picture, why play is central to early learning worldwide
This isn’t just an academic curiosity confined to a handful of university labs. Play-based learning sits at the centre of how major global child development organisations think about early education.
UNICEF and the LEGO Foundation have jointly described learning through play as central to quality early childhood pedagogy, and have pushed for it to be built into early education systems, not treated as a break from “real” learning.[4] Their guidance frames play as one of the most effective ways young children build early mathematical, scientific, and literacy understanding, precisely because it engages children on terms they already find motivating.
This global framing matters because it reflects the same pattern found in the individual research studies. Whether the setting is a Carnegie Mellon lab, a Head Start centre, or a UNICEF-supported program in a country working to close learning gaps, the underlying idea holds up, structured play is not a lesser substitute for direct instruction in early maths, it’s one of the more effective ways to deliver it.
Why games seem to work better than drills for young children
There are a few reasons this keeps showing up across such different research settings.
- Repetition feels natural rather than tedious, because the child wants another turn
- Immediate, built-in feedback, moving too far or too few spaces is obvious right away, without an adult needing to correct the child
- Numbers become something to use for a purpose, comparing scores, moving forward, sharing pieces, rather than something recited in isolation
- Games allow adults to naturally model number talk, “you’re on 7, so you’ll land on 9,” without it feeling like a lesson
There’s also a mindset angle worth mentioning. Stanford mathematics education researcher Jo Boaler has written extensively about how early, low-pressure, exploratory approaches to maths help children avoid the maths anxiety that often develops once timed tests and drills are introduced, an anxiety that can follow children well into their school years. Games, by their playful and repeatable nature, sit naturally on the low-pressure end of that spectrum.
What this looks like as an everyday game at home
None of this requires special equipment. The Head Start study itself used a piece of paper with ten numbered squares drawn on it and a couple of buttons as tokens.
Simple games with strong research behind the format
- A numbered path board game, move a token along numbered squares while counting each one aloud as you move, exactly the format from the original study
- Card games comparing two numbers, flip two cards and ask which is bigger, a simple version of the “war” style card game
- Dice games where the number rolled determines how many spaces to move, steps to take, or objects to collect
- Simple race-to-a-number games, first to collect ten counters, or first to reach the number 20
The common thread in all of these is that the number is doing something. It determines a real outcome in the game, rather than being named and then forgotten.
A simple homemade number path game
This is a close copy of the exact game format used in the original research, and it takes about five minutes to make.
Number Path Game
Materials:
- A strip of paper or cardboard
- A marker
- A die
- Two small tokens, buttons, or coins
How to prepare it:
- Draw ten to twelve squares in a row along the strip.
- Number each square in order, starting from 1.
How to play:
- Each player starts their token before square 1.
- Take turns rolling the die and moving that many squares forward.
- As you move, count each square out loud, starting from the number your token is currently on, “6, 7, 8” rather than starting back at 1 each time.
- First player to reach the final square wins.
That “counting on” step, counting forward from the current position rather than restarting from one, was actually the specific mechanic researchers used in the original study, and it’s thought to be part of why the game worked as well as it did, since it repeatedly reinforces where a number sits relative to the ones around it.[1]
Where this leaves printable and paper-based math games
Printable games can absolutely fit into this picture, provided the maths stays built into the mechanic rather than sitting on the page as a separate task. A printable path game, a roll-and-cover number grid, or a simple card matching set can carry the same research-backed structure as the board game study, just in a print-and-play format rather than store-bought.
Where printables tend to fall short is when the “game” element is really just a worksheet with a game-like theme, colour by number, for instance, is enjoyable, but it isn’t doing the same cognitive work as a game where a wrong count changes the outcome.
What to keep in mind if a game doesn’t seem to be working
Given the home-based card game study found mixed results depending on how parents played alongside their children, it’s worth remembering that a game’s design is only part of the picture.[2] Narrating what’s happening, “you rolled a 4, so you’re moving from 6 to 10,” and giving a child space to count and decide for themselves rather than jumping in with the answer, both seem to matter as much as which game is on the table.
If a particular game doesn’t seem to be landing, it’s rarely a sign the child isn’t ready for maths. It’s more often a sign that particular format isn’t holding their interest yet, worth trying a different mechanic, dice instead of cards, movement instead of a board, before assuming the concept itself is the problem.
References
- Ramani, G. B., & Siegler, R. S. (2008). Promoting Broad and Stable Improvements in Low-Income Children’s Numerical Knowledge Through Playing Number Board Games. Child Development, 79(2), 375–394.
- Ramani, G. B., & Scalise, N. R. It’s more than just fun and games: Play-based mathematics activities for Head Start families. Early Childhood Research Quarterly.
- Clements, D. H., & Sarama, J. (2008). Experimental evaluation of the effects of a research-based preschool mathematics curriculum. American Educational Research Journal, 45(2), 443–494.
- UNICEF & The LEGO Foundation. Learning through Play: Strengthening Learning Through Play in Early Childhood Education Programmes.
- Alotaibi, M. S. (2024). Game-based learning in early childhood education: A systematic review and meta-analysis. Frontiers in Psychology, 15, 1307881.
- Gasteiger, H., Obersteiner, A., & Reiss, K. Formal and informal learning environments: Using games to support early numeracy. In Contemporary Research and Perspectives on Early Childhood Mathematics Education.