Multivariate evolutionary analysis: integrating structural equation modeling and phylogenetics (Visiting Scholar)

One statistical approach with a long history in the social sciences is a multivariate method called Structural Equation Modeling (SEM). Most biological systems are multivariate, which are not easily dissected into their component parts. However, most biologists use only univariate statistical methods, which have definitive limitations in accounting for more than a few variables simultaneously. Therefore, the implementation of methodologies like SEM into biological research is necessary. However, SEM cannot be applied directly to most biological datasets or generalized across species because of the hierarchical pattern of evolutionary history (i.e., phylogenetic non-independence). The goals of my project are: (1) the development of a SEM based multivariate comparative analysis for biological datasets while incorporating the phylogeny on model construction and validation, (2) the development of software for this methodology, and (3) the implementation and generalization of the method using published empirical datasets. The proposed project will provide the foundations for the implementation of the already established SEM-based multivariate techniques into comparative biology. The background of my NESCent project is associated with the development of statistical methodologies for my PhD thesis. During my graduate studies, I became interested in an integrative study of the biology of the poison frog family (Dendrobatidae). My field and lab experiments addressed different aspects of the natural history of this lineage. They include aspects of their evolutionary history (phylogeny and biogeography), physiology (metabolic rates), bioacoustics (advertisement calls), and aposematism (toxicity and conspicuousness). At the end of such experiences, I realized that the available comparative methodologies did not provide an adequate multivariate statistical framework and I decided to search for one. My future research plans will combine the development of other multivariate techniques, apply this techniques to other biological systems, and extending our knowledge of the poison frog system by developing more complex models of their biology. More information about my research at http://www.duke.edu/~js324.

Browse all 5 Items

Multivariate evolutionary analysis: integrating structural equation modeling and phylogenetics (Visiting Scholar)

Project

Title:
Multivariate evolutionary analysis: integrating structural equation modeling and phylogenetics (Visiting Scholar)
Permalink:
https://idn.duke.edu/ark:/87924/r4fp7n
Temporal:
2009-2012
Creator:
Santos, Juan
Type:
Collection
Description:
Project
Abstract:
One statistical approach with a long history in the social sciences is a multivariate method called Structural Equation Modeling (SEM). Most biological systems are multivariate, which are not easily dissected into their component parts. However, most biologists use only univariate statistical methods, which have definitive limitations in accounting for more than a few variables simultaneously. Therefore, the implementation of methodologies like SEM into biological research is necessary. However, SEM cannot be applied directly to most biological datasets or generalized across species because of the hierarchical pattern of evolutionary history (i.e., phylogenetic non-independence). The goals of my project are: (1) the development of a SEM based multivariate comparative analysis for biological datasets while incorporating the phylogeny on model construction and validation, (2) the development of software for this methodology, and (3) the implementation and generalization of the method using published empirical datasets. The proposed project will provide the foundations for the implementation of the already established SEM-based multivariate techniques into comparative biology. The background of my NESCent project is associated with the development of statistical methodologies for my PhD thesis. During my graduate studies, I became interested in an integrative study of the biology of the poison frog family (Dendrobatidae). My field and lab experiments addressed different aspects of the natural history of this lineage. They include aspects of their evolutionary history (phylogeny and biogeography), physiology (metabolic rates), bioacoustics (advertisement calls), and aposematism (toxicity and conspicuousness). At the end of such experiences, I realized that the available comparative methodologies did not provide an adequate multivariate statistical framework and I decided to search for one. My future research plans will combine the development of other multivariate techniques, apply this techniques to other biological systems, and extending our knowledge of the poison frog system by developing more complex models of their biology. More information about my research at http://www.duke.edu/~js324.
Identifier:
  • 185
  • duke:276833
  • ark:/87924/r4fp7n
  • fde74856-0043-4434-a3d3-29b55f127f21
Subject: