Counting to fifty is not the achievement it looks like.
Reciting numbers is a verbal routine, learned the way a song is learned. Children usually have it long before the words attach to amounts. When researchers separated the two and looked at which preschool skills predicted maths in Grade 5, counting with understanding was among the strongest predictors — and plain recitation range predicted essentially nothing.
So the child chanting to fifty and the child who stops at six may be in the opposite positions to the ones you would guess.
What is actually happening, in order
This order is well established and children move through it reliably. The timing is not — the ages below are rough population averages, and the spread between normally developing children is very wide. Read the sequence; hold the ages loosely.
1. Reciting the words
· around 2He chants "one, two, three, four" like a rhyme. This is a song, not maths — and that is completely fine. A child who counts to twenty this way may still have no idea what four means.
2. One word, one thing
· about 2½ to 4Pointing at each object as he says each number. Early on the words run ahead of the finger, or he counts the same biscuit twice. Ordinary, and it sorts itself out with practice.
3. Small amounts start to mean something
· about 2½ to 4He learns what "one" means. Months later, "two". Then "three". Genuinely one number at a time, with long gaps — this is much slower than most parents expect, and it is not a sign of anything.
4. The penny drops: the last word is the answer
· usually between 3½ and 5The big one. He understands that when you count a pile and end on "seven", seven is how many there are — and that this works for any pile. Before this, counting is a ritual he performs. After it, counting answers a question.
5. More, less, and which is bigger
· across 3 to 5Comparing two amounts without counting them out. This one matters more than it looks: knowing how numbers sit in order turns out to be among the better predictors of later maths.
6. First adding and taking away
· about 4 to 5On fingers, with objects, small numbers only. He will count everything from the start before he learns to count on from the bigger number. Both are normal stages.
The one moment worth watching for
Ask your child for five buttons from a pile. What he does next tells you roughly where he is — and it takes ten seconds.
A child who grabs an approximate handful has not yet worked out that the last word in a count gives the total. A child who counts deliberately and stops at five has. Both are ordinary; they are just at different points.
Watch it, do not test it
Slip it into cooking — "pass me three cloves", "give me five spoons". Researchers using this task properly find that even small changes in how it is asked can change how a child appears to perform, so a version done at the kitchen table is a conversation, never a measurement. If he is having a bad day, he will look like he knows less than he does. So will any of us.
The two things with real evidence behind them
1. Count what is in front of him — and go past three
Families were recorded at home over several months. The amount of number talk varied wildly — some children heard roughly sixty times more of it than others. Children who heard more knew more about number a year and a half later, after allowing for how much the family talked generally, and for income and education.
A follow-up found two specifics that make this actionable. Countingobjects the child can see and touch was what related to later understanding — counting with nothing to point at did not. And talk about sets of four to ten mattered most, while making up under a tenth of what parents actually said.
In other words: the highest-value thing is also the thing parents do least. Seven rotis. Six steps. Nine bangles.
Being straight about this one: it is a link, not an experiment. The researchers point out themselves that a child who is naturally drawn to numbers may pull more number talk out of his parents rather than the other way round. It costs nothing, so the case for doing it holds either way — but it has not been proved to cause anything.
2. One board game, played one particular way
This is the strongest finding on this page, because it is a proper experiment rather than an observation. 124 preschoolers were randomly given either a numbered board or an identical board with colours. About one hour of play in total, spread over two weeks.
The number-board children got better at recognising written numerals, judging which number was bigger, counting, and placing numbers on a line. Nine weeks later the difference was still there. The colour-board children improved at nothing — so it was not the attention, the adult time or the fun of playing.
Two details carry the whole effect
The line must be straight. On a linear board the distance travelled, the number of moves, the words spoken and the time taken all grow together with the number — which is what builds a feel for size. A circular track, which is what most board games use, loses most of that.
He names the squares, not the moves. On 3, rolling 2, he says "four, five" — not "one, two". This is exactly how Snakes and Ladders is normally played, and it is the difference between the game working and the game being a pleasant way to pass twenty minutes.
Three things to do
In order of how much evidence sits behind them. The first needs nothing at all.
Count What Is In Front of You
No materials, no set-up, no time set aside. Just counting the rotis, the steps and the bangles out loud — and going past three, which is the part almost every parent skips.
The Great Race
Ten squares drawn on paper, a bottle cap and a die. One hour of this game, played one specific way, measurably changed what preschoolers knew about numbers — and the effect was still there two months later.
How Did You See It?
Show a handful of dots for one second, then cover them. The question you ask afterwards matters more than the answer — and it quietly builds the thinking that addition will need.
Seeing six as three and three
Recognising one, two and three at a glance needs no teaching. What can be built is the step after it — seeing six as parts. That is where addition begins, years before anyone writes 3 + 3 = 6.
Worth doing, and free — though the evidence that training this raises later attainment is weaker than for the board game. What research points to is subtler: children who spontaneously notice number in daily life — "look, three crows" — tend to do better at maths years afterwards. The habit of noticing looks more valuable than trained speed, which is good news, because you can model it for free.
What the evidence does not support
Worksheets, flashcards and formal drilling
No study has isolated worksheets or flashcards for two-to-five-year-olds, so anyone claiming they are proven harmful is overstating. What can be said is firmer than it sounds: there is no evidence they help, early academic gains reliably fade — often by more than half within a year of the teaching stopping — and there are better-evidenced things to do with the same twenty minutes.
The "early maths predicts success" headline
You will meet this one in advertising. It comes from a real and large study, which found maths skills at school entry predicted later achievement more strongly than reading or behaviour did. What the advertising leaves out is that the authors say plainly their results say nothing about which curriculum works — and then point towards play-based rather than drill-and-practice approaches. It is a correlation, not a prescription, and certainly not a case for worksheets at four.
Starting formal school very early
India's own national curriculum framework puts it directly: children should engage with counting, sorting, seriation and patterns before numbers in symbolic form and number operations are introduced. The sequence this page describes is not an imported idea — it is what our own framework asks for.
Worth knowing too: the competencies people quote from that framework — counting to 99, place value, money — are goals for the end of the foundational stage, at about age eight. They are routinely misread as targets for a three-year-old, and that misreading is where a great deal of unnecessary pressure comes from.
What a large Indian survey found
Around 37,000 children aged four to eight were assessed across rural districts in 24 states. Children from private schools were far ahead on written sums — a gap of more than 35 percentage points on numerical addition.
Then the same children were given the same arithmetic as everyday spoken problems. After accounting for background, the gap almost disappeared — around three points. The survey's own conclusion was that the advantage was "mostly limited to the more mechanical operations rather than the day-to-day understanding of these operations".
That is the clearest evidence available, from our own country, that starting the written work earlier buys something narrower than it appears to.
You already have the materials
Rangoli and kolam are pattern and symmetry, laid out dot by dot. A garland is a repeating sequence. Cooking is counting and measuring. Sorting dal is classification. Coins are numerals, quantity and comparison. Counting rhymes in any Indian language are the number sequence set to music.
To be honest about it: no study shows that drawing rangoli improves a child's maths, and we are not going to claim otherwise. What is well supported is the underlying content — patterning, sorting, spatial play, counting things that are present. Our households happen to be full of it already, which means none of this requires buying anything.
When to mention it to a doctor
This is not a checklist, and it is not a test. Maths difficulty cannot be reliably identified at this age, and there is no home screen worth trusting. The commonest mistake here is not missing a problem — it is a parent deciding a perfectly normal child is behind.
With that said, these are reasonable things to raise at a routine check-up around four or five:
- No stable count sequence at all — not even a few numbers in a fixed order.
- Cannot match one number word to one object, even for two or three things.
- Cannot tell which of two obviously different piles is bigger — seven sweets against two.
- No progress at all over six months to a year. Standing still says more than starting late.
- Number difficulty alongside other concerns — speech, hearing, attention, movement. On its own at four, number difficulty means very little; together with other things, it is worth a conversation.
Common questions
›My child can count to fifty. Is he good at maths?
Possibly, but counting far is not the evidence. Reciting numbers is a verbal routine — like knowing the words of a song — and children usually learn it well before the words mean anything. The research bears this out: when preschool skills were tested against maths achievement in Grade 5, counting with understanding was a strong predictor while basic rote counting range predicted nothing once the two were separated. The better question is not how far he counts, but whether he knows that the last number he said is the answer.
›Should I buy worksheets or flashcards for my three-year-old?
There is no evidence they help at this age, and better-evidenced things to do with the same twenty minutes. What research does show is that early academic gains tend to fade, often completely, within a year or two. Counting things around the house and playing a simple number board game have more support behind them than any worksheet, and cost nothing.
›What is the single most useful thing I can do?
Count things your child can see and touch, out loud, and go past three. Studies that recorded families at home found the amount of number talk varied enormously — some children heard sixty times more of it than others — and that counting present objects, especially sets of four to ten, was what related to later understanding. Counting with nothing to point at did not. It needs no materials and fits into cooking, dressing and walking to the shop.
›Should I count with my child in English or in our mother tongue?
In whichever language you speak most naturally with him. The central idea — that the last number word tells you how many there are — carries across languages once a child has it; only the words have to be learned again. A parent counting confidently in Gujarati or Marathi is giving their child more than a parent counting awkwardly in English out of a sense of duty.
›My child grabs a random handful when I ask for five. Is something wrong?
No — that is a recognised and normal stage. Children who have not yet worked out that the last counting word gives the total tend to grab an approximate handful for any number beyond the ones they know exactly. Watching whether your child grabs or counts deliberately is genuinely informative, but it is a stage to notice rather than a problem to fix. It resolves, usually somewhere between three and a half and five.
›Is my child behind if he cannot do what the neighbour’s child does?
Almost certainly not. The order these skills arrive in is very consistent; the timing is not. The window for the big step — understanding that the last word is the total — spans roughly eighteen months across normally developing children. Progress over time tells you far more than position at any moment, and comparing children of this age is the fastest route to making everyone miserable, including the child.
When he is ready for school maths
Once counting means something and small amounts are secure, the next thing that matters is place value — understanding that the 3 in 34 means thirty. It is what everything written down afterwards depends on.
Place value, next →