Fitting Percentage of Body Fat to Simple Body Measurements: College Women

Abstract

Percentage of body fat, age, weight, height, and 14 circumference measurements (e.g., waist) are given for 184 women aged 18–25. Body fat, one measure of health, was accurately determined by an underwater weighing technique which requires special equipment and training of the individuals conducting the process. Modeling body fat percentage using multiple regression provides a convenient method of estimating body fat percentage using measures collected using only a measuring tape and a scale. This dataset can be used to show students the utility of multiple regression and to provide practice in model building.

KEYWORDS:

• Best subsets
• Cross-validation
• Forward selection
• Backward elimination
• Least squares
• Multicollinearity
• Multiple regression
• Parsimony

Supplementary Materials

Dataset (tab delimited text file): BodyFat.txt

Variable Descriptions: BodyFatInfo.txt

Acknowledgments

Permission to make the data available electronically was kindly granted by Dr. Jason V. Slack, Utah Valley University. I would also like to thank Jason for visiting with me on questions related to the selection of women in the study and how measurements were taken. Thanks are also due to the referees and the editor. Their comments led to an improved presentation.