Methodology

Unit conversion

Every height on this site is stored internally in centimetres and converted only for display. 1 inch is treated as exactly 2.54cm — the international standard agreed in 1959, not a rounded approximation. Feet-and-inches results are rounded to the nearest half inch, matching how heights are conventionally written; centimetre results are shown to one decimal place.

Height percentiles (growth charts)

The percentile calculator uses the LMS method, the standard statistical technique behind modern growth charts: a height is converted to a z-score via z = ((X/M)^L − 1) / (L·S), using published L (power), M (median) and S (coefficient of variation) parameters for the person's age and sex, then that z-score is converted to a percentile using the normal distribution.

Two reference datasets are offered, because they're built from different populations and can give a different answer for the same height — see Height percentiles explained for parents for a worked example of the gap between them:

• WHO 2007 Growth Reference — an international reference covering ages 5-19, used as the default outside the US.
• CDC/NCHS — a US-sample reference covering ages 2-20 plus an adult (20+) aggregate point, from Fryar et al.'s 2011-2014 NHANES anthropometric data (2016).

Full citations for both are on the Sources page.

Child height predictor

The predictor uses the mid-parental method, a long-standing estimation formula: for a boy, (father's height + mother's height + 13cm) / 2; for a girl, (father's height + mother's height − 13cm) / 2. The result is shown as a range of ±8.5cm around that midpoint, reflecting the method's typical margin of error.

This is explicitly an estimate based on population statistics, not a medical prediction — actual adult height depends on nutrition, health, puberty timing and other factors the formula can't see. See How tall will my child be? for how this compares to other prediction methods (doubling age 2, Khamis-Roche, bone-age X-rays) and their relative accuracy.

BMI calculator

BMI is weight in kilograms divided by height in metres squared, rounded to one decimal place, with the World Health Organization's adult categories (under 18.5 underweight, 18.5 to 24.9 healthy, 25 to 29.9 overweight, 30 and over obese). The healthy weight range shown is the weight that gives a BMI of 18.5 to 24.9 at the height entered. It is for adults aged 18 and over only, it is an estimate that cannot tell muscle from fat, and the NHS notes that people of Asian, Black African, African-Caribbean and Middle Eastern backgrounds can face higher health risk at a lower BMI. It is not medical advice.

Sitting height ratio

The ratio is sitting height divided by standing height, as a percentage, and estimated leg length is standing height minus sitting height. The tool is descriptive only: there are no clinical cut-offs for a "good" ratio, adult averages differ between populations, and children have a higher ratio than adults because legs grow faster than the trunk before puberty (Bogin and Varela-Silva, 2010). Ratios outside roughly 45 to 62% are flagged as a probable mis-measurement.

You vs the world map and group photo layout

The map colours each country in our dataset by your height minus that country's NCD-RisC mean adult height for the sex you choose; a gap of 1 cm or less counts as the same, and the strongest colour is a gap of 15 cm or more. Countries not in the dataset are grey. The group photo layout on the comparison board places the tallest person in the middle with shorter people stepping down towards the edges and drawn in front, and draws a dashed line at the group's average height. All heights still stand on one baseline at one scale, so the layout changes the arrangement but never the proportions.

Average height by country

Country averages come from the NCD Risk Factor Collaboration's (NCD-RisC) 2019 cohort (mean adult height at age 19), as compiled by World Population Review and heightcomparisoncalculator.com. NCD-RisC's own raw dataset covers roughly 200 countries; this site currently ships a 69-country subset, cross-checked between the two compilations where they overlap. The "average" shown when ranking across this site's own dataset is a simple (unweighted) average of the countries included here — it is not a population-weighted global figure, and leans toward the countries in this particular subset. Full expansion to NCD-RisC's complete country list is a planned follow-up once its data-download page is reachable directly from this build environment.

The comparison board

Subjects on the board share one baseline and one scale: the board's height maps to the tallest subject's height plus 10% headroom, and every silhouette's width scales with its height using that asset's own aspect ratio, so nothing looks stretched or squashed. The ruler automatically picks a step size (from a fixed list such as 1, 2, 5, 10, 20, 50cm, and so on) that gives at least six gridlines, whatever the scale — from a cat next to a skyscraper down to two people a few centimetres apart.

The "in shoes" toggle adds a fixed offset on top of a subject's stored barefoot height: +2.5cm for trainers, +4cm for boots, or a user-set 3-12cm for heels. The underlying stored height is always barefoot; the shod height is only ever used for display and the difference readout notes which heights it's comparing.

A shared board link encodes the comparison itself — names, heights, silhouette choices, colours — directly in the URL. Nothing is stored on a server, and no account is needed to create or share one. An uploaded image is processed entirely in your browser and is never included in a shared link; a link shared from a board with an uploaded image falls back to a generic silhouette for the person who opens it.

Illustrations and figures

The first 50 landmark, object, vehicle and generic-person silhouettes on the board are drawn in code for this site. The remaining silhouettes (people, animals, dinosaurs, plants and further objects, vehicles and landmarks) are AI-assisted: we write the prompts, generate the images with an AI image tool, review each by eye and trace it into a single-colour vector shape. Animal and dinosaur heights are usually quoted to the shoulder or hip, so the board can draw those silhouettes with that point on the height you give; where that point sits on each outline is estimated from the artwork, so treat it as approximate. Nothing is copied, traced or scraped from another height comparison site. Landmark and object heights (for example, a specific building or vehicle) are individually sourced and recorded in this site's own asset registry, with that source noted against the figure wherever it's used in an article.

No real person's photo or likeness appears anywhere on the site. Where a named individual's height is discussed (for example, a verified record height), it's represented as a generic labelled figure at that height, not an image or drawing of them.

Review and updates

Every guide shows a "last reviewed" date. Country-average and growth-reference data are checked against their primary sources periodically; if a figure changes upstream, the article and underlying dataset are both updated and the review date moves forward.

None of the calculators on this site provide medical advice. The percentile and child predictor tools are statistical estimates for general interest and comparison, clearly labelled as such — for any concern about a child's growth or your own health, speak to a GP or paediatrician.

See the full list of primary and secondary sources used across the site on the Sources page.