This preview shows page 1. Sign up to view the full content.
Unformatted text preview: the ordination methods, correspondence analysis was the least used, and multivariate analysis
of variance (MANOVA) was the least used among discrimination
methods. This article provides a brief overview of discriminant
analysis, MANOVA, correspondence analysis, and canonical correlation techniques.
Discriminant analysis. The goal of discriminant analysis is to
assess the association between a single nonmetric dependent variable with two or more levels and a set of several metric independent variables. The procedure has two goals. First, it allows
one to describe and unravel the underlying structure of the association and to determine the contribution of each independent variable in the makeup of that structure. Second, it enables prediction
or classification of new cases based on previously established
structure. Discriminant analysis has been used extensively in the
phytopathological literature to describe the relationship of physical and biological variables to components of disease epidemics.
View Full Document
This document was uploaded on 11/20/2013.
- Summer '11
- The American, Multivariate statistics, MANOVA