Submitted to: Book Chapter
Publication Type: Book / Chapter
Publication Acceptance Date: November 3, 2004
Publication Date: July 1, 2006
Citation: Jaradat, A.A. 2006. Multivariate analyses procedures: Applications in plant breeding, genetics and agronomy. In: Acquaah, G., editor. Principles of Plant Genetics and Breeding. Williston, VT: Blackwell Publishing. p. 155-160. Interpretive Summary: Multivariate analyses (MVAs) procedures, which simultaneously analyze numerous variables measured on individual plants, are increasingly being used in the analysis of plant breeding, genetics and agronomy data. Generally, these procedures are classified as data reduction or data classification methods. We present the salient features and applications of a number of MVAs in plant breeding, genetics and agronomy research. This review would benefit students and scientists in gaining practical knowledge of the benefits and limitations of applying each MVA to their data.
Technical Abstract: Plant breeders, geneticists and agronomists are increasingly faced with theoretical and practical questions of multivariate nature. With increases in germplasm sizes, number of plant and crop variables and evaluation and characterization data on molecular, biochemical, morphological and agronomic traits, multivariate statistical analysis (MVAs) methods are receiving increasing interest and assuming considerable significance. Some MVAs, such as multivariate analysis of variance (MANOVA) and co-variance (MANCOVA), are extensions of uni- and bivariate statistical methods appropriate for significance tests of statistical hypotheses; however, most MVAs are used for data exploration, extraction of fundamental components of large data sets, discovery of latent structural relationships, visualization and description of biological patterns. This review focuses on the salient features and applications of MVAs in multivariate data analyses of plant breeding, genetics and agronomy data. These include MANOVA and MANCOVA, general and mixed regression models for hierarchical and multivariate data analysis, data reduction methods (factor, principal components, perceptual mapping and correspondence analyses), data classification methods (discriminant analysis, clustering and additive trees, conjoint analysis and multidimensional scaling) and categorical data analysis.