Similarities and differences in the univariate and multivariate analysis of repeated measures designs are discussed, using a hypothetical data set studying the effects of practice on the algebra performance of four students to illustrate both methods. When data are analyzed through the univariate approach and the homogeneity assumption is…

A probability distribution was developed for the Andrews-Pregibon (AP) statistic. The statistic, developed by D. F. Andrews and D. Pregibon (1978), identifies multivariate outliers. It is a ratio of the determinant of the data matrix with an observation deleted to the determinant of the entire data matrix. Although the AP statistic has been used…

Canonical correlation analysis is illustrated and three common fallacious interpretation practices are described. Simply, canonical correlation is an example of the bivariate case. Like all parametric methods, it involves the creation of synthetic scores for each person. It presumes at least two predictor variables and at least two criterion…

Homogeneity analysis, or multiple correspondence analysis, is usually applied to k separate variables. In this paper, it is applied to sets of variables by using sums within sets. The resulting technique is referred to as OVERALS. It uses the notion of optimal scaling, with transformations that can be multiple or single. The single transformations…

A brief historical background of discriminant analysis is given, with a description of the variety of roles that discriminant analysis can perform. Focus is on the classification role of discriminant analysis and how it can be performed by using Fisher's classification functions or the canonical discriminant functions. A small hypothetical data…

Conventional canonical methods distinguish between the two variable sets being analyzed, but the methods do not attempt to optimize the variance from a given variable set that will be contained in the final solution. In this respect canonical methods are said the be "symmetric." This paper proposes two non-symmetric, canonical-like…

A particular strategy for investigating effects resulting from a multivariate analysis of variance (MANOVA) is proposed. The strategy involves multiple two-group multivariate analyses. The two groups result from considering multivariate pairwise group contrasts or multivariate complex group contrasts. Assuming a given two-group analysis yields…

The purposes of this study are: (1) to demonstrate the superiority of simple ridge regression over ordinary least squares regression through theoretical argument and empirical example; (2) to modify ridge regression through use of the variance normalization criterion; and (3) to demonstrate the superiority of simple ridge regression based on the…

Multivariate analysis techniques employ various methods of optimally weighting variables to form composites. The interpretation of these composites in subsequent analyses is rarely straightforward, fraught with difficulties based upon the statistical properites of the estimated weights. This paper synthesizes interpretive guidelines for four…

In their Monte Carlo study of canonical analysis, Barcikowski and Stevens evaluated the relative stability of canonical weights and loadings. This paper identifies some weaknesses in their study, suggests directions for future research in this area, and discusses interpretation of canonical analysis both in development and in cross-validation. For…

This article is a rejoinder to TM 502 249. Each of Thorndike's comments are examined. A possible solution to the large number of subjects necessary for stable weights and variate-variable correlations using ridge regression procedures is suggested. (RC)

Presents and illustrates structural equation modeling (multivariate analysis with latent variables, also called causal modeling or covariance structure analysis), discussing issues and problems related to the use of this methodology, possible applications of structural equation modeling to counseling psychology research, and resources for further…

A new approach to partialling components is used. Like conventional partialling, this approach orthogonalizes variables by partitioning the scores or observations. Unlike conventional partialling, it yields a common component and two unique components. (Author/GDC)

Employs latent partition analysis (a multivariate technique) to reveal pre/postcognitive structure of freshman engineering students (N=70) taking general chemistry. Similarities and dissimilarities among high and low scorers are discussed. Arguments for use of this statistical techniques are presented. (DH)

Examined data from a questionnaire concerning people with chronic health, emotional, mental, physical, and sensory impairments, and four respondent demographic and experiential characteristics to determine if 150 human service providers endorsed the premise that components of society disable people who are impaired. Results confirmed that…