Canonical correlation analysis is the most general linear model subsuming all other univariate and multivariate cases (N. Kerlinger & E. Pedhazur, 1973; B. Thompson, 1985, 1991). Because "reality" is a complex place, a multivariate analysis such as canonical correlation analysis is demanded to match the research design. The purpose…

Monte Carlo methods were used to investigate the effects of removing extreme data points identified by five indices of influence. Multivariate normal data were simulated and observations were removed from samples if they exceeded the criteria suggested in the literature for each influence statistic. Factors included in the design of the Monte…

Many multivariate statistical methods call upon the assumption of multivariate normality. However, many applied researchers fail to test this assumption. This omission could be due to ignorance of the existence of tests of multivariate normality or confusion about which test to use. Although at least 50 tests of multivariate normality exist,…

Multivariate techniques have been implemented with greater and greater frequency. In order to use multivariate techniques researchers must understand the fundamental assumptions. The purpose of this paper is to evaluate one of the assumptions of multivariate analysis, normality. Overall, normal distributions are unimodal and symmetrical, and they…

The use of generalized discriminant analysis as a descriptive technique which can be employed outside of the traditional analysis of variance studies is discussed. Examples based on real data are provided. (Author/JKS)

Three potential applications of stepwise procedures in canonical analysis and several alternative stepping decision rules are described. Results of an empirical investigation of the procedures indicated that more parsimonious approaches to maintaining variables held up better under cross-validation. (Author/JKS)

The triangular decomposition method is suggested as a general technique for obtaining the various measures of an ill-conditioned matrix. The advantages of using triangular decomposition are computing nicety, cost, and parsimony. (Author/PN)

The index of redundancy is a measure of association between a set of independent variables and a set of dependent variables. Properties and interpretations of redundancy variables, in a particular subset of the original variables, are discussed. (JKS)

Recent work (EJ 208 813) showing that generalized eigenvalue problems in which both matrices are singular can be solved by reducing them to similar problems of smaller order is discussed. Possible extensions of the work are indicated. (Author/JKS)

This paper discusses considerations in the unwary use of packaged discriminant analysis procedures including: the differences between the "group classification function" and the textbook classification function in both form and use, classification table confusions and their alleviation, and the hazards of stepping procedures. (Author/BW)

To investigate whether research studies on the mildly handicapped have univariate or multivariate research designs, 105 empirical research articles from 1975 to 1979 were reviewed. Factors considered were sample size, type of study, and type of inferential statistic. (SB)

Examples are given to show how the semistrandardized (SS) regression coefficient provides information not given by the conventional standardized regression coefficients used in factor, canonical, and path analysis. (Author/RL)

The purpose of this paper is to outline the role of orthogonal rotation in canonical analysis, including the evaluative measures that need be reported and scrutinized upon application. (Author)

An extension of Wollenberg's redundancy analysis (an alternative to canonical correlation) is proposed to derive Y-variates corresponding to the optimal X-variates. These variates are maximally correlated with the given X-variates, and depending upon the standardization chosen they also have certain properties of orthogonality. (Author/JKS)

Means of identifying sources of rejection of hypotheses regarding linear multivariate statistical models are discussed. Problems with the use of a global test using Roy's largest root criterion and means of solving them are presented, along with a practical application of the techniques. (TJH)