Quantifying asymmetry: ratios and alternatives

Area of Science:

• Anthropometry
• Bioengineering
• Human Anatomy

Background:

• Metric skeletal asymmetry studies traditionally use univariate analyses and ratio transformations to control for size.
• This
• controlling for size
• approach aims to create size-free asymmetry indices by dividing raw asymmetries by a size marker.
• However, ratio transformations often violate the assumption of isometry, potentially obscuring true asymmetries and producing spurious results.

Purpose of the Study:

• To evaluate the effectiveness of classic ratio-based indices in detecting and characterizing asymmetry patterns in four human appendicular bones.
• To explore and propose alternative methodological approaches for analyzing skeletal asymmetry.

Main Methods:

• The study examined the performance of traditional ratio indices on four human appendicular bones.
• Exploratory multivariate techniques, including Principal Components Analysis (PCA).
• Confirmatory linear methods such as regression and analysis of covariance were employed.

Main Results:

• The classic ratio model was found to be inadequate for controlling size in the examined bones, as numerator and denominator variables were independent.
• Ratio transformations introduced significant biases, leading to spurious asymmetry patterns when comparing different elements and variables.
• Multivariate analyses indicated that size and range transformations should precede asymmetry calculations.

Conclusions:

• The traditional ratio-based method for analyzing skeletal asymmetry is unreliable and can generate misleading results.
• Multivariate techniques, combined with linear methods, provide a more robust and accurate alternative for assessing skeletal asymmetry.
• Size and range adjustments should be implemented prior to calculating asymmetry indices for more valid comparisons.