The interpretation of discriminant functions as a follow-up to a significant multivariate analysis of variance is discussed. New indices are proposed that aid in identification and interpretation of the subset of response variables that contribute to a significant group discrimination. Their efficacy is compared to several commonly used…

Interpretations of relative variable importance in multivariate analysis of variance are discussed, with attention to (1) latent construct definition; (2) linear discriminant function scores; and (3) grouping variable effects. Two numerical ranking methods are proposed and compared by the bootstrap approach using two real data sets. (SLD)

A Monte Carlo simulation study compared the power and Type I errors of the Wilks lambda statistic and the statistic of M. L. Puri and P. K. Sen (1971) on transformed data in a one-way multivariate analysis of variance. Preferred test procedures, based on robustness and power, are discussed. (SLD)

This repeated measures factorial design study compared the results of two procedures for identifying multivariate outliers under varying conditions, the Mahalanobis distance and the Andrews-Pregibon statistic. Results were analyzed for the total number of outliers identified and number of false outliers identified. Simulated data were limited to…

The contribution of individual variables to overall multivariate significance in a multivariate analysis of variance (MANOVA) is investigated using a combination of canonical discriminant analysis and Roy-Bose simultaneous confidence intervals. Difficulties with this procedure are discussed, and its advantages are illustrated using examples based…

L-scaling is introduced as a technique for determining the weights in weighted averages or scaled scores for T joint observations on K variables. The technique is so named because of its formal resemblance to the Leontief matrix of mathematical economics. L-scaling is compared to several widely-used procedures for data reduction, and the…

Across conditions, differences between population parameters defined by common factor analysis and component analysis are demonstrated. Implications for data analytic and theoretical issues related to choice of analytic model are discussed. Results suggest that principal components analysis should not be used to obtain parameters reflecting latent…

Multivariate and univariate analysis of variance methods (MANOVA and ANOVA, respectively) are compared for their relative value in educational research. The favoritism shown multivariate techniques is questioned. Criteria for the selection of the appropriate technique are outlined. The relationships among research hypotheses, statistical…

The representation of multivariate data by icons is discussed. The factorial sun is suggested as superior to the commonly used snowflake or sun icons and as better representing the values of the different variables and their correlational structure. Two experiments with 60 college students demonstrate the factorial sun's superiority. (SLD)

Two recent indices of fit, the Relative Noncentrality Index (RNI) (R. P. McDonald and H. W. Marsh, 1990) and the Comparative Fit Index (P. M. Bentler, 1990), are shown to be algebraically equivalent in most applications, although one condition in which the RNI may be advantageous for model comparison is identified. (SLD)

Relationships among individual- and group-directed measures of leader behavior descriptions and five variables were examined using 54 law enforcement agency personnel associated with a large public university. Data on a questionnaire completed by participants during an interview were studied. Explicit consideration was given to multiple levels of…

Complex decision-making environments often demand that a range of alternative strategies be identified and compared for their shared value in achieving desired ends. Simple parametric statistical procedures such as analysis of variance and discriminant function analysis can be used in such situations to compare (using selected criteria) the value…

A bivariate generalization of the genotype-environment covariation (GEC) is presented. A multivariate procedure for detecting univariate and bivariate GEC is also described and illustrated via a study of 136 adopted and 125 non-adopted 4-year-old children. (SLD)

A component analytic method for analyzing multivariate longitudinal data is presented that does not make strong assumptions about the structure of the data. Central to the method are the facts that components are derived as linear composites of the observed or manifest variables and that the components must provide an adequate representation of…

The relationship of multiple linear regression to various multivariate statistical techniques is discussed. The importance of the standardized partial regression coefficient (beta weight) in multiple linear regression as it is applied in path, factor, LISREL, and discriminant analyses is emphasized. The multivariate methods discussed in this paper…