Three observers used 3 different calipers (the Lange and two versions of the Harpenden caliper) to measure skinfold thicknesses in 27 men and 23 women on 8 to 9 occasions over a period of about 1 month. The subjects were aged between 17 and 22 years and none was obese. Measurements were made at the biceps, triceps, subscapular and suprailiac sites. When considering the group as a whole no observer found a significant difference between the two sides of the body for measurements of “total skinfold.” Significant differences were however found by all observers at the triceps site, and by at least one observer at all of the other single sites. For individual subjects there were occasions when one observer found a significant difference in the “total skinfold” between the two sides of the body, but such a difference was rarely found by two observers in the same subject and never by all three. For individual observers the best reproducibility of measurement was found for the “total skinfold”; for single sites, the best were the subscapular in the men and the subscapular and triceps in the women. Worst reproducibility was observed at the suprailiac site in the men and at the biceps and suprailiac sites in the women. The maximum differences between observers—that is the difference between the lowest reading by any observer and the highest reading by any observer—would lead to maximum differences in estimation of fat content of 5% of the body weight for men and 6% for women. No relationship between skinfold thickness and the menstrual cycle could be detected.
Womersley, J and Durnin, J V.G.A
"An Experimental Study on Variability of Measurements of Skinfold Thickness on Young Adults,"
2, Article 12.
Available at: https://digitalcommons.wayne.edu/humbiol/vol45/iss2/12