Child growth
Height-for-Age Percentile Calculator
Enter sex, age, and length or height. Infants use WHO length-for-age (lying down). From age 2 the tool uses CDC stature-for-age (standing). The result is a percentile and z-score, not a predicted adult height.
Reviewed September 2026
Enter your details
Results
Length-for-age and stature-for-age are not the same stick
Under 24 months this page uses WHO recumbent length-for-age. From 24 months it uses CDC standing stature-for-age. Recumbent length is measured lying down. Stature is measured standing. National survey data put standing height about 0.8 cm (a quarter inch) shorter than recumbent length.
If you stand a 20-month-old against a doorframe because you do not have a length board, you are feeding a length chart a stature number. The percentile will read a little low. Clinics that must use standing height on an infant sometimes add 0.7 cm before WHO software. This parent tool does not auto-add. Measure lying down for infants, standing for age 2 and up.
Height-for-age is the linear-growth chart. It is how WHO defines stunting (z below -2) and severe stunting (z below -3). CDC-adapted WHO plots used in US infant clinics mark short stature under the 2nd percentile, which is the same neighborhood as -2 SD.
How to use the height-for-age calculator
Enter sex, completed years, extra months, and the length or height from the same day. US units take inches. Metric takes centimeters. The highlighted percentile is the comparison with the active chart (WHO or CDC).
Hair puffs, braids stacked high, and shoes all add false stature. Remove them. The Frankfort plane (ear canal to lower orbit) should be level. Heels, buttocks, and shoulders against the board or wall.
The result is not a predicted adult height. Mid-parental height equations live in clinic, not on this page.
Type 4 feet 2 inches as 50 inches, not 4.2. The field is inches, not feet. 4.2 inches is a newborn length and will print a severe-stunting reading for an 8-year-old.
If the child will not stand still, sit them out and try after a snack. A leaning stance shortens stature. So does bent knees. One good standing try beats three crouched ones averaged.
Formula and which table loads
Inches convert at 2.54 cm per inch. Z = ((cm / M)^L - 1) / (L x S). Percentile follows the normal CDF. WHO months 0 to 24. CDC 24 to 240 months at the half-month CDC age point.
A child who is 24 months old on the nose uses CDC stature-for-age. If you still measured recumbent length at that visit, say so at the appointment. The 0.8 cm method gap can move a child near a cutoff.
Linear measurement this page expects
| Age | Chart | Position | Common home error |
|---|---|---|---|
| 0 to 23 months | WHO length-for-age | Lying on a firm board, head and feet held | Marking a wall while the toddler stands |
| 24 months to 20 years | CDC stature-for-age | Standing, no shoes | Keeping sneakers on "because they always wear them" |
| Past 20 years | None | n/a | Using a child chart for an adult |
Worked example: 8 years, 50 inches
Boy, 8 years, 0 extra months, 50 in (127 cm). CDC stature-for-age medians for boys near 96 months sit a little above 127 cm. Expect a percentile around the middle, not "short" because 50 inches sounds small next to a sixth grader.
Girl, 0 years, 9 months, 28 in (71.1 cm). WHO length-for-age for girls at 9 months has a median near 70 to 71 cm. That measurement should plot near the 50th if the board reading is true recumbent length.
Drop that 9-month girl to 25 in (63.5 cm) and Z falls below -2. The reading prints stunting. That is not a wait-and-see number. Bring length, weight, and head circumference to the same visit.
A 14-year-old girl at 50 in is far below CDC stature-for-age. That is a clinic problem, not a "she takes after grandma" decision you make from this box. Mid-parental height can explain the 10th percentile. It does not explain the 0.1st without an exam.
The 3rd percentile, the 2nd percentile, and velocity
School-age clinicians often treat the 3rd percentile as a short-stature screen on CDC charts. Infant WHO/CDC training language uses the 2nd percentile. This tool prints a CDC size band that calls out below the 3rd after age 2, and WHO stunting language before age 2.
One low point is weaker evidence than a flattening velocity. A child who was on the 40th and sits on the 10th a year later with healthy parents who are both 5 feet 10 inches needs a workup more than a child who has always tracked the 5th with two short parents.
Length-board details that change the z-score
Two adults help: one holds the head against the headboard, one straightens the legs and brings the footboard to the heels. A wiggling toddler's first reading is often 1 cm short. Repeat until two readings agree within 0.5 cm and use the longer of the clean pair if the protocol in clinic does that. At home, average two careful tries and do not "stretch" the child.
A 1 cm error at 9 months is a large fraction of a standard deviation on WHO length tables. That is bigger than most month-to-month true growth in late infancy.
Marking the wall with a book on the head is better than a baseball cap brim. It is still worse than a stadiometer. If home and clinic disagree by more than 1 cm, trust the clinic device until you copy their stance.
Infants with hip dysplasia harnesses cannot always lie fully straight. Record that on the visit note. This calculator will still run a number that may not match the orthopedics chart.
What to pair with a height percentile
Always read weight-for-age or weight-for-length next. Isolated tall stature with delayed weight is a different clinic question than isolated short stature with high weight-for-length.
After age 2, BMI-for-age uses this height. A 1 inch overread of stature lowers BMI and can hide overweight. Measure height as carefully as you weigh.
Constitutional delay of growth and puberty is a clinic diagnosis that often shows delayed bone age plus family history of late growth spurts. A single low stature percentile at 8 years is not that diagnosis.
Tall stature with headaches, vision changes, or very large hands is a different call than tall stature in two 6-foot parents. Bring the height number to the visit. Do not skip the other symptoms because the calculator printed "very tall."
Accuracy limits, edge cases, and safety
Scoliosis, contractures, and hypotonia make stature a poor number. Clinics may use arm span or sitting height. This page cannot.
Achondroplasia and other skeletal conditions need condition-specific charts. A CDC or WHO percentile will look extremely low and is the wrong conversation.
Corrected age for prematurity applies to length the same way it applies to weight in the first two years. Calendar age on this tool will make a former 28-week infant look short until you correct, which is expected and should be done with the pediatrician.
Do not start a stretching device, growth-hormone inquiry on a forum, or calorie restriction to "slow height" from one home measurement.
Errors that move height-for-age more than a month of growth
| Error | Typical size | Who it hurts most | Fix |
|---|---|---|---|
| Shoes left on | 1.5 to 3 cm | School-age CDC stature | Bare feet |
| Standing an infant | About 0.8 cm short vs length | WHO length-for-age near a cutoff | Recumbent board |
| Ponytail under a stadiometer | 0.5 to 2 cm | Girls with high buns | Take the bun down |
| Using last year's height | Several cm after a spurt | BMI-for-age more than this page | Remeasure today |
Puberty timing changes the stature channel
CDC stature-for-age is a cross-section by completed month, not a bone-age chart. An early-maturing 10-year-old can sit high on stature-for-age and later join peers. A late-maturing 14-year-old can sit low and still reach a mid-parental target. This page cannot see Tanner stage.
Girls' pubertal growth spurt usually starts earlier than boys'. Comparing a 12-year-old girl at the 75th with a 12-year-old boy at the 40th is not a fair "who is taller for their age" kitchen argument. They are on different sex tables and often different pubertal clocks.
If height velocity from two accurate visits a year apart is near zero in a school-age child, that matters more than whether this calculator printed the 12th or the 18th percentile today. Bring the two clinic heights, not two doorframe pencil marks taken in different shoes.
Endocrine referral thresholds are clinic policy. Some groups use height below -2 SD plus slow velocity. Some use the 3rd percentile plus mid-parental mismatch. This tool will not schedule that referral. It will give you the percentile to take in.
How it works
Z = ((cm/M)^L - 1) / (L x S). Measure infants lying down on a length board. Measure children 2 and older standing, without shoes. Mixing those methods by a centimeter or more can move the percentile.
Frequently asked questions
Standing height on a 20-month-old?
WHO length charts assume recumbent length. Standing height is usually a little shorter. Ask the clinic which number they plotted.
Is below the 3rd percentile short stature?
It is a common screen. Diagnosis needs mid-parental height, growth velocity, and a clinician, not one calculator visit.
Related calculators
References
- WHO. Length/height-for-age
- CDC NCHS. WHO Growth Charts Data files
- CDC. Using WHO Growth Standard Charts
- CDC NCHS. CDC Growth Charts Data Files (LMS values)
- CDC MMWR. Use of World Health Organization and CDC Growth Charts for Children Aged 0-59 Months in the United States
- Hampl et al., AAP. Clinical Practice Guideline for the Evaluation and Treatment of Children and Adolescents With Obesity
- AAP. Bright Futures
- CDC. Growth Charts
Kitchen tools are estimates for home cooking. Nutrition tools are screening aids, not a diagnosis. They are not a substitute for a pediatrician or registered dietitian. Food-safety temperatures follow USDA FSIS. Do not reuse leftover meat brine.