How to Predict a Child's Height as an Adult (Mid-Parental Method)
Updated: 2026-10-30
The single best number you can get for a child's adult height comes from the two parents, because height is about 80% heritable. The classic mid-parental formula is one line: take the parents' midpoint and shift it toward the child's sex by 6.5 cm. It is honest about being a family average, not a measurement — which is why it ships with a wide ±8.5 cm band rather than a false-precise figure. This guide shows the formula exactly as you type it, the difference between it and the Khamis-Roche method people search for, and when a browser tool beats doing it by hand.
The short answer
Take the mother's and father's heights in centimetres, average them, and add or subtract 6.5 cm for the child's sex. For a boy, (mother + father) ÷ 2 + 6.5 cm. For a girl, (mother + father) ÷ 2 − 6.5 cm. That is the mid-parental estimate, the same formula growth clinics use as a first pass.
The 6.5 cm shift exists only because adult men are on average about 13 cm taller than adult women, so you move the parents' midpoint by half that difference in the direction of the child's sex. Round it, and the answer is what you would expect from a family of that stature — not a promise of a specific adult.
- Boy: (mother + father) ÷ 2 + 6.5 cm
- Girl: (mother + father) ÷ 2 − 6.5 cm
- Example, 160 cm mother and 178 cm father, a boy: (160 + 178) ÷ 2 + 6.5 = 180.5 cm
- Same parents, a girl: 169 − 6.5 = 162.5 cm
- The tool shows a range of estimate ± 8.5 cm on top of the point figure
Why the answer is a band, not a number
Twin and family studies put the heritability of adult height around 80% in well-nourished populations. The remaining 20% is everything a formula cannot see: childhood nutrition, sleep, chronic illness, and the timing of puberty. A child whose growth plates close earlier or later than average can land at either end of the band for reasons that have nothing to do with the parents' heights.
That is why the honest output is a range. The ±8.5 cm band on this tool is deliberately wide — it is a statement that most children end up within a few inches of the family average, not that the middle number is the one they will hit. If a source quotes your child's future height to the nearest centimetre, it is presenting an estimate as a measurement, and you should treat it as the former.
How to calculate it from your family's heights
You need two inputs: the mother's height and the father's height, plus the child's sex. That is all the mid-parental method uses. The child's own current age and height are not part of the formula — they are shown on the tool purely as context, so you can see where the child sits relative to the prediction, but they never change the number.
Work in centimetres first. If you only have feet and inches, convert: 1 inch is 2.54 cm, so multiply feet by 30.48 and inches by 2.54 and add. The tool handles both unit systems for you and displays the result in both, so you do not have to do the conversion yourself.
Step by step
Add the two parents' heights together in centimetres.
Divide by two to get the family midpoint.
Add 6.5 cm for a boy, subtract 6.5 cm for a girl.
Read the result as a point estimate, and use the ±8.5 cm range as the realistic spread.
A worked example
Mother 165 cm, father 182 cm. Midpoint is (165 + 182) ÷ 2 = 173.5 cm. A boy comes out at 173.5 + 6.5 = 180 cm, with a realistic band of roughly 171.5 to 188.5 cm. A girl comes out at 173.5 − 6.5 = 167 cm, band roughly 158.5 to 175.5 cm.
Notice the girl's estimate is not just the midpoint minus 6.5 — it is the same midpoint, shifted down by the sex difference. The two children from the same parents differ by 13 cm, which is exactly the average adult sex difference the formula is built around.
Mid-parental versus the Khamis-Roche method
Searches for "height predictor khamis-roche" are the most common way this tool shows up, so the difference is worth being precise about. Khamis-Roche is a more refined statistical method: it uses the child's own current height and weight (for children roughly five and up) together with validated percentile tables to project adult height, and it typically narrows the uncertainty somewhat.
This tool does not run Khamis-Roche. It runs the mid-parental method, which needs only the parents' heights and the child's sex, and which works in the opposite direction — parents' heights in, child's future out. It cannot predict for an adult who is already done growing, and it does not use a bone-age X-ray. If you want the tighter, child-data-based projection, that is a method that needs the child's measurements and published tables, not a parents-only calculator.
Boys versus girls: where the 6.5 cm comes from
The 6.5 cm shift is half of the average adult difference between men and women, which is about 13 cm. The parents' midpoint is a family average that ignores the child's sex, so the formula nudges it halfway toward the sex-appropriate end. That is why two children of the same parents, one a boy and one a girl, come out 13 cm apart in the estimate — not because boys grow more from the same genes, but because the endpoint the formula targets is different for each sex.
This is also why the answer moves by the full 13 cm when you change the sex, rather than by a small rounding difference. The shift is built from the population average, so it applies to the typical family; a family where the mother is tall and the father is short simply has a different midpoint to shift, and the boy and girl from it still land 13 cm apart.
If you do not have both parents' heights
The mid-parental formula needs both parents. If you only have one, you have no reliable substitute — averaging one parent with a population mean for the other introduces an assumption the method was never built to make, and the result is less of an estimate and more of a guess.
The honest answer in that case is the growth chart, not a formula: a paediatrician tracks a child's own percentile over time, and staying near their own curve is what matters clinically. Percentiles answer a different question than this prediction does — they describe where the child is on their own trajectory, while the mid-parental estimate guesses the endpoint from the family. Do not force one into the other's place.
When the prediction drifts: growth spurts and timing
The mid-parental number is an endpoint, not a path. Two children can reach the same adult height on completely different schedules: one grows steadily through childhood, the other sits near the bottom of the chart and then catches up in a late growth spurt. Puberty timing is the main driver, and it is influenced by genetics in a way the parents' heights alone do not capture.
If a child's percentile on their own growth chart is drifting upward or downward for a while, that is a question for a paediatrician, not for this calculator. The tool tells you a reasonable endpoint from the family; it does not tell you whether the child is on track for it. Re-running it with the child's current height added changes nothing in the number — that field is context, so it is worth not reading more into it than the method supports.
When to use the calculator instead of doing it by hand
The formula is two lines, so doing it in your head is fine for a one-off. The calculator is worth using when you want the unit conversion and the range handled without a slip, when you are comparing a boy and a girl from the same parents, or when you want a clean, private, offline answer without a web search that does not send your family's numbers anywhere.
It takes the two parents' heights and the child's sex, returns the point estimate in centimetres and feet/inches plus the ±8.5 cm range, and optionally shows the child's current age and height as context. Nothing you type is stored or sent to a server — the whole thing runs in your browser.
- Enter both parents' heights in cm or feet/inches and choose boy or girl.
- Read the point estimate and the ±8.5 cm range.
- Optionally enter the child's current age and height to see where they sit relative to the prediction.
- Not a diagnosis and not a legal or medical figure — for concerns about a child's growth, a paediatrician and a growth chart are the right tool.
Frequently asked questions
How accurate is the mid-parental height formula?
Good enough for a family ballpark, not for a specific number. It captures roughly 80% of the variation that comes from genetics, so most children land within a few inches of the estimate, but the other 20% — nutrition, sleep, illness and the timing of puberty — is invisible to it. Treat the result as a range (the tool shows ±8.5 cm), not as a precise centimetre.
What is the Khamis-Roche method and does this tool use it?
Khamis-Roche is a more refined statistical projection that uses the child's own height and weight (for children about five and older) with validated percentile tables, and it can narrow the uncertainty somewhat. This tool does not run it. It runs the mid-parental method, which needs only the parents' heights and the child's sex. It is also not a bone-age assessment (an X-ray that estimates remaining growth).
Can I use it to predict my own adult height as an adult?
No. The method only makes sense for a child whose final height is still ahead of them, and it requires the two parents' heights. If you are already grown, your height is fixed — growth plates have closed regardless of what the formula would output. Estimating an adult's height from a shoe size or a photo is guesswork by comparison.
Does the child's current height matter?
Not in the calculation. The mid-parental formula uses only the parents' heights and the child's sex. The optional fields for the child's current age and height on the tool are context only — they let you see how the child compares to the prediction, but they never change the number.
What if I only know one parent's height?
There is no reliable mid-parental substitute with one parent. Averaging one parent with a population mean for the other adds an assumption the method was not built for. In that situation a paediatrician's growth chart, which tracks the child's own percentile over time, is the more honest route than forcing a formula.
How is height inherited?
Height is strongly genetic — heritability is around 80% in well-nourished populations — but it is not a single gene. Dozens of variants each shift the outcome a little, which is why siblings with the same parents can differ by several centimetres. The mid-parental formula approximates the genetic midpoint and then shifts it by the sex difference; it does not model the individual variants.
At what age does a child usually reach their adult height?
Most children finish the bulk of their growth by the end of puberty, which for girls is often around 14 to 16 and for boys a couple of years later. The growth plates close over in that window, and once they have, no formula, diet, or exercise adds height. The mid-parental estimate predicts that endpoint; it does not predict the age at which it is reached.
Is there a free tool that does this in the browser without sending my data?
Yes. The height predictor on this page takes the two parents' heights and the child's sex, and returns the mid-parental estimate in centimetres and feet/inches plus a ±8.5 cm range. It runs entirely in your browser, stores nothing, and sends no network request. It is a family-average estimate, not a medical prediction, so use it as a first pass and a growth chart for anything that needs to be clinical.