Skinfold Body Fat Calculator

A skinfold caliper measures a fold of skin and underlying subcutaneous tissue at a specified location. Published equations relate the sum of selected skinfolds and age to estimated whole-body density. A separate density conversion produces an estimated body-fat percentage.

These equations were derived from particular study populations. The result is an estimate, not a direct measurement of all body fat, a diagnosis or a fitness grade. This calculator does not attach athletic, obesity or health-risk labels to the percentage.

Select the method before entering measurements. Every required skinfold must be a finite positive value in millimeters. A missing site is not treated as zero, and an abdominal or midaxillary measurement cannot replace biceps in the four-site method.

Jackson–Pollock 3-site

Male equation: chest + abdominal + thigh; ages 18–61. Female equation: triceps + suprailiac + thigh; ages 18–55. Let S be the sum in millimeters and A be age in whole years:

Male density = 1.10938 − 0.0008267S + 0.0000016S² − 0.0002574A.

Female density = 1.0994921 − 0.0009929S + 0.0000023S² − 0.0001392A.

Jackson–Pollock 7-site

For both equations, S = chest + abdominal + thigh + triceps + subscapular + suprailiac + midaxillary. Male ages 18–61; female ages 18–55.

Male density = 1.112 − 0.00043499S + 0.00000055S² − 0.00028826A.

Female density = 1.097 − 0.00046971S + 0.00000056S² − 0.00012828A.

Durnin–Womersley 4-site

S = biceps + triceps + subscapular + suprailiac for both sexes. Density = c − m × log10(S), using the age and sex coefficients below from Table 5 of the 1974 paper.

The pooled Durnin study included 481 people aged 16–72, but Table 1 separates 209 men aged 17–72 and 272 women aged 16–68. This calculator enforces those sex-specific age limits. It does not use a separate adolescent equation below them.

The male Jackson–Pollock derivation included 308 men and cross-validation on another 95 (November 1978). The female derivation included 249 women and cross-validation on another 82 (1980). The original female paper advises extra caution above age 40. Being within an age range does not prove individual accuracy.

Use a trained measurer and the landmarks specified by the selected protocol. A skinfold is not a circumference, and labels for nearby hip or trunk sites are not interchangeable. Keep the caliper, measurer, side and protocol consistent when comparing repeated estimates.

Durnin and Womersley usually measured the right side with the subject standing relaxed. Their subscapular fold was approximately 45° to vertical, and their suprailiac site was just above the iliac crest at the midaxillary line. These published details matter when selecting a matching measurement.

The Temple et al. validation study (online April 16, 2014; issue February 2015) measured the right side, raised a double fold of skin and tissue, and applied the caliper at right angles 1 cm below the pinch. Sites were measured in rotating order at least twice. If two readings differed by more than 2 mm, a third was taken; the study used the mean of two readings or the median of three.

Those are details of a documented study procedure, not a substitute for training in locating every site. Enter measurements from the selected method; if the available measurements do not match, obtain the correct sites rather than substituting a convenient location.

All modes use the Siri conversion: body fat (%) = 495 / density (g/mL) − 450. Estimated fat mass = weight × body fat (%) / 100; estimated fat-free mass = weight − fat mass. Fat-free mass includes bone, organs and water as well as muscle.

For a 25-year-old man with biceps 5, triceps 10, subscapular 12 and suprailiac 15 mm, Durnin S = 42 mm. Density = 1.1631 − 0.0632 × log10(42) = 1.0605106 g/mL. Body fat = 16.7563%, displayed as 16.76%. At 80 kg, estimated fat mass is 13.41 kg and fat-free mass is 66.59 kg.

The calculator displays the method, sum and density so the arithmetic can be checked. It rejects nonpositive or nonfinite density and percentages outside 0–100%, rather than clamping the result into a plausible-looking range. Both calculated mass components must be finite, nonnegative and no greater than total weight.

Weight may be entered in kilograms or pounds, using 0.45359237 kg/lb for conversion. Skinfold values remain in millimeters in either weight unit. Any input, sex, method or unit change clears the old estimate.

Caliper measurements are portable, but technique and tissue properties affect the estimate. Body-density conversions also depend on assumptions about the density of fat and fat-free tissue; differences from those assumptions can change the estimate.

Derivation ages describe the available study data. The original Durnin sample was not a random population sample. The calculator rejects ages outside the selected sex-specific derivation range to avoid unmarked age extrapolation. It does not establish suitability in pregnancy, illness or body composition substantially different from the study populations.

Different methods can produce different numbers because their sites and fitted equations differ. More sites do not guarantee a more accurate individual result. The displayed decimals express calculation precision and should not be treated as an individual error margin.

Can I use abdominal skinfold instead of biceps?

No. Durnin–Womersley requires biceps, triceps, subscapular and suprailiac. A different site sum needs its own matching published equation.

Can I calculate outside the stated age range?

This tool does not extrapolate beyond the derivation ages. Obtain professional guidance on an appropriate assessment rather than changing the entered age.

Does the estimate classify health or athletic status?

No. The density equations do not provide the former fitness and health-risk categories.

Primary derivations: Jackson and Pollock, Generalized equations for predicting body density of men (November 1978); Jackson, Pollock and Ward, Generalized equations for predicting body density of women (1980); Durnin and Womersley, measurements on 481 men and women (July 1974), Tables 1 and 5.

Coefficient and protocol checks: Temple et al. (2014/2015), Table 1 and testing procedure and Shakibaee et al. (December 1, 2015), Table 1 male JP3/JP7 equations.

Sources and methods checked: 2026-09-30.