Unformatted text preview: u of discriminant analysis, logistic regression may be used
in modeling the relationship of a binary dependent variable (yes
or no, presence or absence) to metric or nonmetric independent
variables. As such, logistic regression may be viewed as a discriminant analysis with two levels of the dependent variable and
with independent variables that are either metric or nonmetric.
Press and Wilson (15) studied the performance of discriminant
analysis and logistic regression, and concluded that if the assumption of multivariate normality of independent variables is violated,
logistic regression is preferable to discriminant analysis.
MANOVA. MANOVA is a procedure for assessing differences
among several nonmetric independent variables based on the
linear combination of several metric dependent variables. This
procedure enables the simultaneous examination of several
dependent variables. Golinski et al. (6) used MANOVA to assess
the effect of two pathogens (Fusarium avenaceum and F. culmorum) on three yield components (1,000-kernel weight and
weight and number of kernels per winter wheat head) of 14
winter wheat cultivars in a 2-year study. The de...
View Full Document
- Summer '11
- The American, Multivariate statistics, MANOVA