Premium
A COMPARISON OF MULTIVARIATE METHODS FOR THE DETECTION OF HYBRIDIZATION
Author(s) -
Adams Robert P.
Publication year - 1982
Publication title -
taxon
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.819
H-Index - 81
eISSN - 1996-8175
pISSN - 0040-0262
DOI - 10.2307/1219680
Subject(s) - multivariate statistics , multivariate analysis , statistics , biology , mathematics
Summary Principal component analysis (PCA), Wells' hybrid distance diagram, principal coordinate analysis (PCO) using weighted Gower metrics, and canonical variate analysis (CVA) were compared using data from artificial crosses of sunfish ( Lepomis ) and putative hybridization in Juniperus . The results were similar between the two data sets in that using the weighted characters (F ratios from ANOVA of the parents) produced well resolved parental groups and generally ordinated the hybrids between the parents. The F weighted PCO and Wells' distance diagrams were found to be better suited for the analysis of hybridization than PCA or CVA. Several limitations of PCA and CVA are discussed.