ABOUT / METHODS

Which skinfold method should you use?

Skinfold equations estimate body density or body-fat percentage from measurements at selected sites. Because each method was developed for a particular population and protocol, the most relevant option is generally the one that best matches the person being measured and the way the measurements are taken.

COMPARISON

Methods in this calculator

1978 men / 1980 women

Jackson–Pollock 3-site

Study sample and evidence

  • Men: 18–61 years; 308 in the development sample and 95 in cross-validation.
  • Women: 18–55 years; 249 in development and 82 in cross-validation.
  • The studies covered a range of ages and body-fat levels.
  • The authors advised caution when applying the women's equations above age 40.

Sites

Men: chest, abdomen, thigh. Women: triceps, suprailiac, thigh.

Formula

Men BD = 1.10938 − 0.0008267Σ₃ + 0.0000016Σ₃² − 0.0002574·age. Women BD = 1.0994921 − 0.0009929Σ₃ + 0.0000023Σ₃² − 0.0001392·age. %BF = Siri(BD).

Strengths

Three sites make this method relatively quick. The equations were developed for adults across a range of ages and body-fat levels, then cross-validated in separate samples.

Limitations

Accurate results depend on consistent site-specific technique and selecting the right age- and sex-specific equation. The original studies do not establish equal accuracy across ethnicities, athletes, older adults or people with very different body compositions.

Studies and sources

Jackson & Pollock (1978), men · Jackson, Pollock & Ward (1980), women

Calculator implementation

Jackson–Pollock 4-site

Study sample and evidence

  • This calculator uses sex-specific equations to estimate body-fat percentage from four sites: abdomen, suprailiac, thigh and triceps.
  • This variant differs from the original generalized Jackson–Pollock 3- and 7-site body-density equations.
  • The cited original studies do not establish this exact 4-site version as a separate gold-standard method.

Sites

Abdomen, suprailiac, thigh, triceps.

Formula

Men: %BF = 0.29288Σ₄ − 0.0005Σ₄² + 0.15845·age − 5.76377. Women: %BF = 0.29669Σ₄ − 0.00043Σ₄² + 0.02963·age + 1.4072.

Strengths

Uses four familiar measurement sites and directly estimates body-fat percentage.

Limitations

The target population and external validation for this specific equation are less clearly documented than for the classic Jackson–Pollock 3- and 7-site methods. Treat it as an additional calculator option, not as the best-validated Jackson–Pollock method by default.

Studies and sources

Jackson & Pollock (1978), men · Jackson, Pollock & Ward (1980), women

1978 men / 1980 women

Jackson–Pollock 7-site

Study sample and evidence

  • Uses the same broad adult samples as the Jackson–Pollock 3-site equations.
  • Men: 18–61 years; women: 18–55 years.
  • The women's study advised caution for women over 40.

Sites

Chest, midaxillary, subscapular, triceps, abdomen, suprailiac, thigh.

Formula

Men BD = 1.112 − 0.00043499Σ₇ + 0.00000055Σ₇² − 0.00028826·age. Women BD = 1.097 − 0.00046971Σ₇ + 0.00000056Σ₇² − 0.00012828·age. %BF = Siri(BD).

Strengths

Measures more sites across the trunk and limbs and is supported by the classic generalized Jackson–Pollock studies.

Limitations

Takes longer and creates more opportunities for site or measurement error. More sites do not automatically produce a more accurate estimate for an individual.

Studies and sources

Jackson & Pollock (1978), men · Jackson, Pollock & Ward (1980), women

1974

Durnin–Womersley 4-site

Study sample and evidence

  • 481 men and women from Glasgow: 209 men and 272 women.
  • Participants were aged 16–72.
  • Body density was measured by underwater weighing.
  • The sample covered a broad range of body-fat values.
  • Equations are separated by sex and age band.

Sites

Biceps, triceps, subscapular, suprailiac.

Formula

BD = a − b·log₁₀(Σ₄); %BF = Siri(BD). Both a and b depend on sex and age band.

Strengths

Age-banded equations and a wide adult age span; only four sites. A reasonable starting point when the person resembles the study’s adult population and the protocol is followed.

Limitations

Historical UK sample; the oldest bracket is 50+ in the calculator despite the original sample ending at 72. It may misestimate people with substantially different body-fat distribution or composition.

Studies and sources

Durnin & Womersley (1974) · Siri (1961), body-density conversion

Source history uncertain

Parrillo 9-site

Study sample and evidence

  • The calculator uses this rule: Σ₉ (mm) × 27 ÷ weight (lb).
  • A peer-reviewed derivation or validated population for this exact equation has not been identified.
  • It is therefore unclear which population the equation best fits.

Sites

Chest, subscapular, abdomen, suprailiac, triceps, biceps, thigh, calf and lower back; body weight is also required.

Formula

%BF = Σ₉ × 27 ÷ body weight (lb).

Strengths

Covers several body regions and is straightforward to calculate when all nine measurements and body weight are available.

Limitations

Evidence for this exact equation and its target population is unclear. It also requires more measurements, and its result should not be treated as interchangeable with estimates from validated body-density methods.

Studies and sources

No reliable original source has been identified for this specific equation.

PRACTICAL GUIDE

How to choose

  • For a classic generalized equation for adults, consider Jackson–Pollock 3-site or 7-site when the person's age, sex and measurement protocol fit the published studies.
  • If a four-site method and age-specific equations suit your needs, consider Durnin–Womersley, particularly when the person falls within the original adult age range.
  • Jackson–Pollock 4-site and Parrillo are also available, but the evidence for these specific versions is less clearly established. Treat them as additional estimates for comparison.
  • For children, pregnancy, clinical decisions or people who differ substantially from the original study samples, seek a validated method for that population and guidance from a qualified professional.

ATHLETE NOTE

Athletes with low body fat

No single formula is best for every lean athlete. Look for a method validated against a multi-compartment reference in athletes with a similar sport, sex, age and body-fat range. Fitness research does exist: a validation study compared 21 equations with underwater weighing in 265 male athletes and found that only three Jackson–Pollock equations were not significantly different from the reference average; results still varied by equation. A separate study of white female NCAA Division I athletes found no improvement from adding sites beyond three. Neither study validates every skinfold formula or proves that seven sites are always better than three. To monitor training-related changes, track the skinfold sum (Σ in mm) over time using the same protocol and, if possible, the same assessor. A very low estimate alone is not a performance or health goal.

3 sites

Faster and simpler to repeat. A good practical choice when you want a consistent trend and can reliably measure the method's specified sites. In this calculator, JP3 uses different sites for men and women.

7 sites

Takes longer and depends on accurate measurement at more sites. It samples more body regions, but that alone does not guarantee a better body-fat estimate. Choose it when you can follow its full protocol consistently or need continuity with a 7-site assessment.

EQUATIONS

Formula notes

Σ is the sum of skinfold thicknesses in millimetres; age is in years; BD is body density in g/cm³. Siri conversion: %BF = (495 ÷ BD) − 450. This two-compartment conversion assumes fixed densities of fat and fat-free mass and can introduce additional error when that assumption does not fit. JP coefficients and the implemented JP4 equations are reproduced from this calculator’s source; Durnin–Womersley uses log base 10. Parrillo uses pounds for weight, even if weight is entered in kilograms in the calculator.

TIMELINE

A short history

  1. 1950s–60s

    Early equations linked skinfolds to body density in relatively narrow samples. Siri’s 1961 conversion provided a widely used way to turn density into a two-compartment body-fat estimate.

  2. 1974

    Durnin and Womersley published age- and sex-specific equations from 481 adults aged 16–72, using four skinfolds and underwater-weighing body density.

  3. 1978

    Jackson and Pollock developed and cross-validated generalized equations for 308 and 95 men aged 18–61, with different site-count options.

  4. 1980

    Jackson, Pollock and Ward extended generalized 3-, 4- and 7-site equations to women aged 18–55 and validated them in a second sample.

  5. Later practice

    The equations remain common in field and fitness settings, but validation reviews show that accuracy varies across equations and populations. A formula’s familiar name does not make it universal.

FAQ

Frequently asked questions

Which skinfold formula is best for estimating body fat?

For a practical starting point, choose Jackson–Pollock 3-site. If you want to measure more body regions and can repeat the longer protocol consistently, choose the 7-site version. Durnin–Womersley 4-site is another option when its age-specific protocol fits. The calculator’s Jackson–Pollock 4-site and Parrillo 9-site options have less clearly documented evidence.

Why do skinfold formulas give different body-fat results?

They use different measurement sites, equations and study populations. Technique and site placement also affect the measurements. Each result is an estimate, so differences between formulas are expected.

Is Jackson–Pollock 7-site more accurate than 3-site for body-fat measurement?

Not automatically. Seven sites sample more body regions and take longer to measure, but more sites alone do not guarantee a more accurate estimate for an individual. Consistent technique and a suitable equation matter.

Which skinfold formula is suitable for lean athletes?

No single formula is best for every lean athlete. Prefer a method validated in athletes with a similar sport, age, sex and body-fat range. For tracking change, also compare the skinfold sum using the same protocol and assessor.

How accurate is body-fat measurement with skinfold calipers?

Accuracy varies. Skinfold equations estimate body-fat percentage, and results can be affected by the caliper, measurement technique, age, body-fat distribution and how closely a person matches the study population. They are not a diagnosis.