A variety of problems associated with the interpretation of traditional canonical correlation are discussed. A response surface approach is developed which allows for investigation of changes in the coefficients while maintaining an optimum canonical correlation value. Also, a discrete or constrained canonical correlation method is presented. (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)

A procedure is presented to transform an n-dimensional scatter diagram into a two-dimensional scatter diagram while preserving proximity relationships between points. This procedure can help in the presentation and interpretation of multivariate results. (Author/SLD)

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…

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)

Methods of regression commonality analysis are generalized for use in canonical correlation analysis. An actual data set (involving educators' attitudes toward death and age, locus of control, religion, and occupational role in working with terminally ill children) is employed to illustrate the extension. The method can be applied with respect to…

In the present study Monte Carlo methods were employed to evaluate the degree to which canonical function and structure coefficients may be differentially sensitive to sampling error. Sampling error influences were investigated across variations in variable and sample (n) sizes, and across variations in average within-set correlation sizes and in…

Multiple regression analysis is frequently being employed in experimental and non-experimental research. However, when data include predictor variables that are correlated, some regression results can become difficult to interpret. This paper presents a study to provide a demonstration that structure coefficients may be useful in these cases.…

The test proposed by Brien, Venables, and Mayo (1984), endorsed by Silver and Dunlap, and supported by computer software that they developed is not a proper test for multivariate independence. A revised appraisal is suggested for the Silver and Dunlap results. (SLD)

The need for using invariance procedures to establish the external validity or generalizability of statistical results has been well documented. Invariance analysis is a tool that can be used to establish confidence in the replicability of research findings. Several approaches to invariance analysis are available that are broadly applicable across…

Since computer programs have been available for estimating and testing linear causal models, these models have been used increasingly in the behavioral sciences. This paper discusses the problem that very different causal structures may fit the same set of data equally well. (Author/LMO)

This paper discusses the principles underlying discriminant analysis and constructs a simulated data set to illustrate its methods. Discriminant analysis is a multivariate technique for identifying the best combination of variables to maximally discriminate between groups. Discriminant functions are established on existing groups and used to…

Multivariate methods were used to identify between-set factors relating the criterion set of eleven Wechsler Adult Intelligence Scale Revised subtest variables to the predictor set of demographic variables: age, race, sex, education, occupation, geographic region, and urban versus rural residence. Although factor analysis is usually used to…

Canonical correlation analysis is a multivariate statistical model which facilitates the study of interrelationships among multiple dependent variables and multiple independent variables. It identifies components of one set of variables that are most highly related linearly to the components of the other set of variables. The underlying logic of…

The tests for zero partial association in a multiple contingency table have gained new importance with the introduction of graphical models. It is shown how these may be performed as exact conditional tests, using as test criteria either the ordinary likelihood ratio, the standard x squared statistic, or any other appropriate statistics. A…