Abstract
In anthropometric design, one usually tries to accommodate the widest range of sizes from small to large. The accelerating ratio of physical size to percentile as one approaches the upper extreme end of the Gaussian distribution, with its generally concomitant cost and/or spatial impacts on the physical system, can often limit the maximum percentile or minimum percentile which can be realistically accommodated. However, what is not generally realized is the relative sizes of the physical dimensions involved because so much work is done with percentiles which are dimensionless quantities. This paper presents several figures which provide some idea of the relative physical sizes of percentiles. The rather small coefficient of variance for anthropometric measurements means that the difference between say the 50th and the 95th percentiles while being a difference of “45 percentile points” is only a matter of a few percent in physical size. Such considerations can help in providing a better understanding of the tradeoff in percentiles and percent and perhaps adds some physical interpretations of percentiles.
Get full access to this article
View all access options for this article.
