Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Scaling, normalizing, and per ratio standards: an allometric modeling approach

A M Nevill1, R L Holder

  • 1School of Sport and Exercise Sciences, University of Birmingham, United Kingdom.

Journal of Applied Physiology (Bethesda, Md. : 1985)
|September 1, 1995
PubMed
Summary

Allometric modeling offers a superior approach to normalizing physiological data compared to traditional ratio standards or linear regression. This method effectively addresses statistical issues like heteroscedasticity and skewness, providing more accurate body size adjustments.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

BMI is dead; long live waist-circumference indices: But which index should we choose to predict cardio-metabolic risk?

Nutrition, metabolism, and cardiovascular diseases : NMCD·2022
Same author

Common mental disorders among Irish jockeys: prevalence and risk factors.

The Physician and sportsmedicine·2020
Same author

Modeling children's development in gross motor coordination reveals key modifiable determinants. An allometric approach.

Scandinavian journal of medicine & science in sports·2018
Same author

Scaling waist girth for differences in body size reveals a new improved index associated with cardiometabolic risk.

Scandinavian journal of medicine & science in sports·2016
Same author

The need to redefine age- and gender-specific overweight and obese body mass index cutoff points.

Nutrition & diabetes·2015
Same author

Physical employment standards for U.K. fire and rescue service personnel.

Occupational medicine (Oxford, England)·2015

Area of Science:

  • Physiology
  • Biostatistics
  • Human Performance

Background:

  • Physiological variables are often normalized using "per ratio" standards (Y/X).
  • Linear regression is proposed as an alternative, but relies on assumptions of constant variance and normal distribution of residuals.
  • These assumptions are frequently violated for physiological data, such as maximum oxygen uptake and power output.

Purpose of the Study:

  • To propose and evaluate allometric modeling as an alternative to "per ratio" standards and linear regression for normalizing physiological variables.
  • To demonstrate how allometric models inherently address statistical challenges like heteroscedasticity and skewness.
  • To show the appropriateness of allometric or log-linear models for predicting dependent variables using normalized physiological data.

Main Methods:

Related Experiment Videos

  • Utilized allometric modeling, where the ratio concept is integral to the model structure.
  • Applied multiple regression analysis with log-transformed dependent and independent variables.
  • Incorporated log-transformed "per ratio" variables as independent predictors in the regression models.

Main Results:

  • Allometric models naturally overcome heteroscedasticity and skewness issues common with "per ratio" variables.
  • Log-linear models, derived from allometric principles, are more suitable than linear models for incorporating "per ratio" standards into predictive regression.
  • The proposed multiple regression approach automatically identifies the most appropriate "per ratio" standard based on the allometric model.

Conclusions:

  • Allometric modeling provides a statistically robust framework for normalizing physiological data.
  • This approach is superior to traditional "per ratio" standards and standard linear regression when assumptions are violated.
  • Log-linear multiple regression using transformed "per ratio" variables offers an effective method for accurate physiological data adjustment and prediction.