Synthesizing modern child and infant anthropometry for design

Open Access
Pagano, Brian Thomas
Graduate Program:
Mechanical Engineering
Master of Science
Document Type:
Master Thesis
Date of Defense:
Committee Members:
  • Matthew B Parkinson, Thesis Advisor
  • anthropometry
  • child
  • infant
  • anthropometry synthesis
  • human variability
  • principal component analysis
  • obesity
  • ergonomics
The most recent comprehensive study of child (2.00 to 18.99 years of age) and infant (birth to 1.99 years of age) anthropometry (body size) was performed by Snyder et al. in 1977 at the University of Michigan. For over 30 years, this dataset has been the primary source of anthropometry in the design of products, environments, and safety standards for infants and children. Since 1977, the United States has seen large increases in the prevalence of overweight and obesity, compromising the distributions of anthropometric measurements in infants and children and slowly decreasing the accuracy of the 1977 dataset. This research aims to synthesize (predict) accurate datasets of modern child and infant anthropometry for use in design. Relationships between anthropometric variables were successfully extracted from the Snyder et al. datasets through the use of principal component analysis, data weighting based on demographics, and other statistical techniques, and were applied to predictor anthropometry from the National Health and Nutrition Examination Survey (NHANES) dataset. The methodology developed for this work generates detailed anthropometry for thousands of stature and body mass index pairs for children and infants in NHANES from 1999-2008 to create a virtual population. Synthesized data were visualized using various methods, including percentile curve estimates. Results were compared to anthropometry from NHANES, the Centers for Disease Control and Prevention growth charts, and the original Snyder et al. dataset. After comparing the synthesized modern virtual population to the 1977 data, measures of width and circumference in children were observed to have undergone large increases in the upper percentiles while lower percentiles increased very little. Measurements of the torso and legs experienced much larger changes than those of the head, face, hands, and feet. Measures of length in children showed no consistent change. Anthropometric measures in infants increased in measures of length, width, and circumference with larger increases observed starting in the later half of the second year of age. Two case studies were also performed in which the consequences of differences in accommodation between the Snyder et al. dataset and new virtual population were explored for bicycle helmets and child safety seats.