Adult height prediction
A family-based adult height estimate from both parents’ measurements.
- Method
- Mid-parental height formula
- Result
- Estimate and family target range
Find the calculation, reference data and limits behind each tool. Start with the one you use.
A family-based adult height estimate from both parents’ measurements.
A current height compared with children of the same age and sex.
The mid-parental method gives a family-based estimate using both biological parents’ heights and a sex-specific adjustment.
The formula uses parents’ heights in centimeters. It does not use your child’s current height, weight, age or bone age.
Boys: (mother’s height + father’s height + 13 cm) ÷ 2
Girls: (mother’s height + father’s height − 13 cm) ÷ 2
For a mother measuring 160 cm and a father measuring 180 cm, the estimate is 176.5 cm for a boy or 163.5 cm for a girl.
We show a family target range of 8.5 cm below and above the estimate, following Children’s Mercy’s guidance. For the boy in the example, that is 168–185 cm. It is a broad family reference, not an individually validated confidence interval or a promise about a particular child.
Puberty timing, health, nutrition and measurement accuracy can affect eventual height. This formula cannot tell you how much growth a child has left or whether growth plates have closed.
All calculations use centimeters internally. Imperial inputs convert at exactly 2.54 cm per inch and use the same 13 cm adjustment. Switching units does not change the formula; a decimal place does not mean the estimate is accurate to a tenth of a centimeter.
Both parents’ heights are required. We do not substitute a population average for an unknown height. You can use the height percentile calculator to understand a current measurement instead.
A percentile describes a current measurement’s position in the CDC reference. It does not predict adult height.
We use the CDC 2000 stature-for-age reference for standing height, for ages 2 years to under 20 years. The comparison uses the child’s age at measurement and the male or female reference curve.
This is a historical U.S. population reference, not a worldwide adult height distribution. Infant length and adult comparisons require different references; this tool does not extrapolate outside its age range.
We use age- and sex-specific LMS parameters: L (a transformation parameter), M (the median) and S (a variation parameter). Completed months select the CDC month interval. For example, 10 years 0 months uses the 120.5-month row, representing ages 120 to under 121 months.
z = ((height / M)L − 1) / (L × S)
When L is zero: z = ln(height / M) / S
The z-score converts to a percentile using the standard normal distribution. Very small or large results display as below 0.1 or above 99.9 rather than an absolute 0 or 100.
The chart shows the 3rd, 50th and 97th percentile reference curves and the measurement you enter. These curves describe the reference population, not your child’s future growth.
The 50th percentile is the median, not an ideal height or a health score. A single percentile cannot diagnose a condition or rule one out. Growth over time and the child’s wider history provide context. See illustrations of the 50th and 75th percentiles for a plain-language explanation.
References sit with the tool they support: adult height prediction and height percentiles. Research on other prediction methods appears in the further reading below.
These publications explain the methods and data. They do not constitute an independent validation of our website, and the institutions listed do not endorse or operate My Height Predictor.
Documentation updated October 13, 2026. Educational information only; our tools do not provide a medical diagnosis or an individualized clinical assessment.
Other adult height prediction methods use different inputs. They are not used by our calculators.
This model uses age, sex, current height, weight and parents’ heights. It was developed using the Fels Longitudinal Study, with age- and sex-specific coefficients. A 1995 erratum corrected tables in the original paper.
Its original study describes limitations to its applicability. The model should not be confused with the CDC percentile reference or applied outside its supported population and age range without appropriate evidence.
Methods such as Bayley–Pinneau and Roche–Wainer–Thissen use assessed skeletal age and other information. They belong in clinical assessment and are not offered here. More inputs do not automatically make a method appropriate for every child.