A technique for estimating anthropometric data from the parameters of the distribution of stature

Publisher Website
Google Scholar
Abstract
A variety of biological and social factors influence the anthropometric characteristics of any population. The ergonomist requires anthropometric data which are appropriate to the target population with whom he is concerned. Since formal surveys are both costly and time consuming, simple methods of estimation are commonly required. The purpose of this paper is to describe and validate one such method. A collection of anthropometric sources dealing with adult European and North American populations was assembled. A selection of dimensions generally useful for the purposes of workspace design was made. The sources were divided into two groups (A and B). For each dimension in each group A source, two ratios were calculated, relating its mean and standard deviation to those of stature. These ratios were then averaged across the group A sources (taking male and female populations separately), to give scaling factors which could be used for predictive purposes. For purposes of validation, the scaling factors derived from group A were used to predict the results of the six published surveys in group B. (Only the sex and the mean and standard deviation of stature of the group B surveys were used in the predictions.) Comparisons were made between the predicted 1st and 99th percentile value of each dimension and the ‘true’ values published in the surveys. Errors were normally distributed with a mean value of —3 mm and a standard deviation of 13 mm (136 comparisons). The actual percentiles which the predicted 1st and 99th percentiles represented were calculated. On average, these included 96% of the population (as against 98% for a perfect prediction) Many of the predictions fell within the confidence limits for the 1st and 99th percentiles of the published surveys. These inaccuracies are modest in comparison with those which may occur, for example, in correcting for clothing, shoes or ‘sitting slump’. It is concluded therefore that anthropometric data estimated by this technique are sufficiently accurate for many practical purposes.

This publication has 8 references indexed in Scilit: