How this instrument works
Mid-parental height is a simple, decades-old way to estimate a child's likely adult height from the two numbers most families already know: the father's and mother's height. The method averages the parents' combined height and adjusts it by a fixed 13cm to account for the average height difference between adult men and women — adding 13cm before halving for a son, subtracting 13cm before halving for a daughter. The result is a single predicted height in centimeters, plus a likely range spanning roughly the 3rd to 97th percentile of outcomes, calculated as the prediction plus or minus 8.5cm.
This is the classic method described by Tanner, Goldstein, and Whitehouse in 1970, built from height-growth data on British children. It is distinct from the more elaborate Khamis-Roche method sometimes used clinically, which additionally requires the child's current height, weight, and age to place them on a growth curve, and tends to narrow the prediction range further. This calculator implements only the simpler mid-parental version, which needs nothing beyond the two parents' heights.
Both versions are population-level statistical estimates, not individual forecasts. A child's actual adult height depends on nutrition, overall health, chronic illness, puberty timing, and other genetic factors beyond the two parents' height alone, so any individual child can land outside even the roughly 3rd-to-97th-percentile range this calculator reports. Pediatricians use tools like this as one data point among several — alongside a child's own growth chart over time — rather than as a definitive prediction, and persistent concerns about a child's growth trajectory are best raised with a pediatrician who can track percentiles over time.
- Set "Predicting a daughter's height (female)" to Yes for a daughter, or leave it at No for a son — the calculator adds 13cm for sons and subtracts 13cm for daughters before halving.
- Enter the father's height in centimeters.
- Enter the mother's height in centimeters.
- Read the predicted adult height in centimeters, plus the likely range spanning roughly the 3rd to 97th percentile.
Worked example — 178cm father, 165cm mother
For a son: (178 + 165 + 13) ÷ 2 = 356 ÷ 2 = 178.0cm predicted, with a likely range of 169.5cm to 186.5cm (178.0 ± 8.5).
For a daughter from the same two parents: (178 + 165 − 13) ÷ 2 = 330 ÷ 2 = 165.0cm predicted, with a likely range of 156.5cm to 173.5cm — noticeably shorter than the son's prediction, reflecting the average height difference the 13cm adjustment is built to capture. For a taller couple — a 185cm father and 170cm mother — a predicted son comes out at (185 + 170 + 13) ÷ 2 = 184.0cm, with a range of 175.5cm to 192.5cm.
Questions
How accurate is mid-parental height prediction?
It's a population-level statistical estimate, not a precise forecast for any one child. The ±8.5cm range is designed to cover roughly the 3rd to 97th percentile of actual outcomes for children with a given mid-parental height, meaning most children land inside it but a meaningful minority don't. Real adult height depends on nutrition, overall childhood health, and other genetic factors beyond just the two parents' height, so treat the result as a ballpark rather than a guarantee.
Why does the formula add or subtract 13cm?
The 13cm adjustment approximates the average height difference between adult men and women. Adding it before halving (for a son) or subtracting it (for a daughter) corrects the simple average of two parents' heights — one male, one female — toward the typical height of the child's own sex, rather than leaving the prediction sitting between the two sexes.
Is this the same as the Khamis-Roche method?
No. This calculator uses the simpler, older mid-parental height method from Tanner, Goldstein, and Whitehouse (1970), which needs only the two parents' heights. Khamis-Roche is a separate, more involved method that also factors in the child's current height, weight, and age, generally producing a tighter prediction range — but it requires data this calculator doesn't collect and isn't what's implemented here.
Does this account for the child's current growth or health?
No — this calculator uses only the two parents' heights and the child's sex. It doesn't know the child's current height, growth rate, puberty stage, or health history, all of which a pediatrician would factor in alongside a formula like this one when actually assessing a child's growth trajectory.
Can this predict height for a child who isn't born yet?
The method only needs both parents' heights, so it can technically be calculated at any point, but it says nothing about factors that affect a specific pregnancy or childhood — nutrition, illness, or other health conditions — that can shift an individual child's actual adult height away from this population-based estimate.
Why is the son's prediction always taller than the daughter's from the same two parents?
Because the formula's only difference between sons and daughters is the fixed 13cm adjustment, reflecting the average adult height gap between men and women. Given identical parent heights, that built-in 13cm difference (before halving, so 6.5cm net) is what separates the two predictions — it isn't a reflection of anything else about the specific parents.
References
Read this first: This instrument computes a screening figure from population formulas — it is not a diagnosis, and it cannot see the whole picture a clinician can. Use it to inform a conversation, not to replace one.