This paper explains in user-friendly terms why multivariate statistics are so important in educational research. The basic logic of canonical correlation analysis is presented as a simple or bivariate Pearson "r" procedure. It is noted that all statistical tests implicitly involve the calculation of least squares weights, and that all…

Some interpretations of relative variable importance in the contexts of multivariate analysis of variance (MANOVA) and discriminant analysis (DA) are presented. Some indices potentially useful for the interpretations are presented, and the assessment of variable importance is illustrated using real data sets. Both descriptive discriminant analysis…

A longitudinal approach is demonstrated that allows assessment of: the means by which a student's level and rate of learning in a given subject area compare with the level and rate of learning of other students, and each student's relative strengths and weaknesses across subject areas. The approach involves an individual growth model to estimate…

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…

The mathematical extension from scalars to matrices as a prerequisite for univariate to multivariate generalizations of statistical concepts and methods is discussed. One reason for several alternative techniques is the multitude of ways to generalize from univariate concepts to multivariate concepts. The additional complexity, and lack of…

The acquisition of information from pictures and the relationship between canonical orientation and the function of objects were studied. Twenty introductory educational psychology students experienced a set of five common objects, three different functional uses, and four possible orientations for each object. The results analyzed for one…

The past decade has seen a substantial growth in methods and schemes for the display of multivariate data. This paper encompasses a sketch of the history of multivariate displays, from the pre-Columbian Quipu to Chernoff's Face; examines a number of techniques; describes their construction; illustrates their use; and comments on their efficacy.…

Prediction for a number of criteria from a set of predictor variables in a system of regression equations is studied with the possibilities of linear transformations applied to both the criterion and predictor variables. Predictive composites representing a battery of predictor variables provide identical estimates of criterion scores as do the…

This paper reviews and develops summary measures of associations between multiple sets of variables through the application of canonical correlation analysis. These measures are subsequently applied to a specific research problem: a study of the determinants of housing status. Some of the data analysis situations for which canonical correlation is…

A computer-simulated study was made of the sampling distribution of omega squared, a measure of strength of relationship in multivariate analysis of variance which had earlier been proposed by the author. It was found that this measure was highly positively biased when the number of variables is large and the sample size is small. A correction…

The partitioned-U procedure is outlined, a fundamental logical flaw in this procedure's avoidance of any direct test of the significance of the first discriminant function or largest coefficient of canonical correlation is pointed out, and two alternatives to the partitioned-U procedure are discussed. (Author/DEP)

Multivariate Nominal Scale Analysis (MNA) was used to assess the independent ability of each variable to predict respondents who indicated a fear of crime (42 percent) and those who did not (58 percent). (Author/AM)

Comparison of three microcomputer statistical packages, SPSS, SAS, and BMDP, focuses on ease of use, ease of learning, documentation, accuracy, similarity to the mainframe version, and speed. The results of eight statistical multivariate analyses used to evaluate their performance are given. (LRW)

Describes the loglinear model, and applies it to categorical data cross-tabulated on four dimensions. Defines contrasts similar to those of the analysis of variance and describes post hoc and planned comparison strategies. Illustrates hypothesis testing and model building for categorical data, providing guidelines for performing an analysis on…

Reparameterization is used to find the maximum likelihood estimates of parameters in a multivariate model having some component variable observable only in polychotomous form. Maximum likelihood estimates are found by a Fletcher Powell algorithm. In addition, the partition maximum likelihood method is proposed and illustrated. (Author/GDC)