All parametric methods are special cases of canonical correlation analysis and can, in fact, be performed using the canonical correlation procedures found in commonly available computer statistics packages such as the Statistical Analysis System (SAS) software. It is suggested that canonical analysis has advantages of versatility since it can…

Planned comparisons have been known for several years. Due to the availability of computers, these comparisons have become a more viable alternative to post hoc testing. There are several different types of planned comparisons that can be performed. Research goals must be well thought out when using planned comparisons, since the appropriate…

Canonical correlation (CC) analysis is discussed with a view toward providing an intuitive understanding of how the technique operates. CC analysis entails calculation of one or more sets of canonical variate coefficients (CVC), i.e., weights which can be applied to the variables in a study. A canonical function (CF) always consists of exactly two…

Designed to provide comparative analyses which could be employed as a basis for judicious program selection and use of generally available canonical correlation computer programs. (RC)

Describes how the test statistic for nonparametric one-way multivariate analysis of variance can be obtained by submitting the data to a packaged computer program. Monte Carlo evidence indicates that the nonparametric approach is advantageous under certain violations of the assumptions of multinormality and homogeneity of covariance matrices.…

When selection occurs on the basis of two or more predictors, multivariate restriction of range can reduce various parameters of a validation study. A Statistical Analysis System (SAS) and a Fortran IV program are described that allow for correction of criterion standard deviation(s) and zero-order validities. (Author)

This paper presents three variable deletion strategies in canonical correlation analysis. All three strategies are illustrated by examples. The first strategy uses the canonical communality (h2) coefficients of the three functions to decide which variable to delete. The second function also uses the canonical communality coefficients, but only…

The stabilities of standardized (ß) and structure (rs) coefficients in canonical (CA) and discriminant analyses (DA) were studied. Four different situations were studied--two pertaining to CA and two to DA. The situations were meant to represent "somewhat typical" and yet varying research conditions that often would not be thought to be…

Data are often a continuous function of a variable such as time observed over some interval. One or more such functions might be observed for each subject. The extension of classical data analytic techniques to such functions is discussed. (Author/JKS)

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)

This paper explains how a major statistical package (BMDP) can be used to produce partial, semipartial, or bipartial set correlation in terms of a procedure outlined by Karpman (1980). (BW)

The canonical redundancy statistic, an estimate of the amount of shared variance between two sets of variables, exhibits an amount of bias similar to that of the first squared canonical correlation coefficient. Two formulae, Wherry and Olkin-Pratt, adequately correct the bias of the redundancy statistic. (Author/BW)

Problems of sampling error and accumulated rounding error in canonical variate analysis are discussed. A new technique is presented which appears to be superior to canonical variate analysis when the ratio of variables to sampling units is greater than one to ten. Examples are presented. (Author/JKS)

Conducted simulations to explore methods for comparing bivariate distributions corresponding to two independent groups, all of which are based on Tukey's "depth," a generalization of the notion of ranks to multivariate data. Discusses steps needed to control Type I error. (SLD)