Child Height Predictor
An estimate of adult height from the parents heights, with the error bars stated honestly rather than hidden behind a confident-looking figure.
In short
The mid-parental method averages the parents heights, then adds 6.5 cm for a boy or subtracts 6.5 cm for a girl. It predicts adult height to within about 8 to 10 cm either way for most children, which is wide enough that it should be treated as a range rather than a number.
Last reviewed
Runs entirely in your browser. Nothing you type here is uploaded, stored or logged.
The mid-parental method
This is the calculation paediatricians actually use, and it is deliberately simple.
boy: (mother + father) ÷ 2 + 6.5 cm
girl: (mother + father) ÷ 2 − 6.5 cmThe 6.5 cm is half the average adult height difference between men and women, which is about 13 cm across most populations. Averaging the parents estimates the family genetic contribution; the adjustment shifts it to the right distribution for the child.
An older version added or subtracted 13 cm to one parent before averaging. It gives the same answer and is more confusing, so the version above has largely replaced it.
Why the range is so wide
Around 95% of children finish within about 10 cm either side of the mid-parental estimate. That is a 20 cm window, which is wide enough that the estimate cannot tell you whether a boy will be 175 cm or 190 cm.
The reason is that height is not inherited as an average. It is polygenic, influenced by hundreds of variants, and a child receives a random half of each parent's set. Two parents of the same height can produce children who differ substantially, and regression toward the population mean means very tall parents tend to have children slightly shorter than themselves, while very short parents tend to have children slightly taller.
This is, in fact, where the statistical term "regression" comes from. Francis Galton coined it in the 1880s while studying exactly this: the tendency of children's heights to fall back toward the average.
What the other 20% is
Within a well-nourished population, roughly 80% of height variation is genetic. The remainder is environment, and it is not evenly distributed.
- Nutrition, particularly protein and micronutrients during infancy and puberty. This is the largest single environmental factor and the reason average heights in many countries rose 10 cm or more over the twentieth century, far too fast for genetics to explain.
- Sleep. Growth hormone is released in pulses, largely during deep sleep. Chronic sleep restriction during the growing years has a measurable effect.
- Chronic illness. Untreated coeliac disease, inflammatory bowel disease, kidney disease and severe asthma can all suppress growth, and treating them often restores it.
- Early-life health. Repeated infections in infancy divert energy from growth, which is much of why height correlates with childhood living conditions at a population level.
Notably absent from that list: specific foods, supplements, stretching and sports. There is no evidence that any of these increase final adult height in a healthy, well-fed child.
Growth patterns matter more than height
A single height measurement tells you very little. The pattern over time tells you almost everything.
Children grow along a percentile line, and the line they are on is far less important than whether they stay on it. A child consistently on the tenth percentile is a small child in a family of small people, and that is not a medical finding. A child who was on the fiftieth at four and the tenth at eight has crossed two major percentile lines, and that is worth investigating.
Rough guidance on growth velocity: about 25 cm in the first year, 12 cm in the second, then 5 to 6 cm a year through childhood, rising to 8 to 12 cm a year at the peak of the pubertal spurt. Less than 4 cm a year between the ages of four and puberty is the usual threshold for a referral.
Puberty timing changes everything temporarily
Girls typically start their growth spurt around 10 or 11 and finish growing between 14 and 16. Boys start around 12 or 13 and finish between 16 and 18.
That two-year offset produces the familiar period between about 11 and 13 when girls in a class are noticeably taller than boys. It reverses completely, and it tells you nothing about final height.
Late developers are the group most often worried about unnecessarily. A boy who is short at 14 because puberty has not started yet has more growing time ahead of him than his peers, and constitutional delay of growth and puberty, which usually runs in families, typically resolves with a normal adult height.
What this cannot do
It does not know the child. It uses two numbers, and it cannot see the growth chart, puberty stage, bone age or health history, all of which a paediatrician would weigh far more heavily.
Treat it as a rough family expectation and nothing more. If there is genuine concern about a child's growth, a growth chart from the last few years is worth vastly more than any formula, and a doctor can read one in about a minute.
Frequently asked questions
How do you predict a child adult height?
The mid-parental method averages the two parents heights, then adds 6.5 cm for a boy or subtracts 6.5 cm for a girl. That figure is the target height, and the child is expected to land within roughly 8 to 10 cm either side of it.
How accurate is it?
Accurate as a range, poor as a number. Around 95% of children finish within about 10 cm either side of the mid-parental estimate, which is a 20 cm window. Anyone quoting a prediction to the centimetre is overstating what the method can do.
How much of height is genetic?
Around 80% within a well-nourished population, spread across hundreds of gene variants rather than a few. The remaining 20% is mostly nutrition, sleep, chronic illness and early-life health, which is why average heights in many countries rose substantially over the twentieth century.
When do children stop growing?
Girls typically finish around 14 to 16, boys around 16 to 18, when the growth plates fuse. Girls start their growth spurt about two years earlier, which is why girls are often taller than boys of the same age between about 11 and 13 before the pattern reverses.
Should I worry if my child is short?
Height on its own matters far less than the growth pattern. A child who has always tracked the tenth percentile is usually just a small child. A child who crosses downward through percentile lines, or grows less than about 4 cm a year between age four and puberty, is worth discussing with a doctor.