Estimators which are optimal under assumptions of normality are shown to be vulnerable to the effects of outliers. A survey of robust alternatives is presented. Included are alternatives to the mean, standard deviation, product-moment correlation, t-test, analysis of variance, multivariate techniques, and schemes for outlier detection. (Author/JKS)

Multiple regression analysis (MRA) may be used when both continuous and categorical variables are included as independent research variables. The use of MRA with categorical variables involves dummy coding, that is, assigning zeros and ones to levels of categorical variables. Caution is urged in results interpretation. (Author/CH)

Some of the more simplified methods for contrasts with equal sample sizes in multiple regression analysis are shown to result in erroneous calculations when applied to data sets with unequal sample sizes. An alternative method is provided. (Author/JKS)

The amount of variance accounted for by treatment can be estimated with omega squared or with the squared multiple correlation coefficient. Monte Carlo methods were employed to compare omega squared, the squared multiple correlation coefficient, and the squared multiple correlation coefficient to which a shrinkage formula had been applied, in…

Regions of significance in aptitude-by-treatment-interaction studies are examined by the traditional statistical approach and an alternative approach which integrates: (1) testing for the parallelism of two or more regression lines; (2) testing for their identity; and (3) Scheffe's theorem. (Author/RL)

Recent criticism in the literature of the use of inferential statistics in educational research is refuted. The authors focus on the defense of multiple regression analysis. (JKS)

Four computer programs using the general purpose multiple linear regression program have been developed. Setwise regression analysis is a stepwise procedure for sets of variables; there will be as many steps as there are sets. Covarmlt allows a solution to the analysis of covariance design with multiple covariates. A third program has three…

Joreskog's general method for analysis of covariance structures was developed for estimating a model involving structures of a very general form on means, variances, and covariances of multivariate observations. This method achieves a great deal of generality and flexibility, in that it is capable of handling most standard statistical models as…

In this report, the author suggests changes in the treatment of overhead costs by hypothesizing that "the effectiveness of standard costing in planning and controlling overhead costs can be increased through the use of probability theory and associated statistical techniques." To test the hypothesis, the author (1) presents an overview of the…

The degree of nonorthogonality in a factorial design was systematically increased. Five methods of dealing with nonorthogonality were selected and applied: two were least squares solutions (Method 1 and Method 2); two were approximate solutions (the unweighted means analysis and the method of expected frequencies); and the fifth was the…

This article discusses M. Rosenberg's Test-Factor Standardization as a type of marginal standardization. To illustrate this relation, the author utilizes a table on abortion attitudes formerly published by K.W. Smith in his commentary on Rosenberg's work. (Author/GC)

Canonical analysis of two convex polyhedral cones involves looking for two vectors whose square cosine is a maximum. New results about the properties of the optimal solution to this problem are presented. The convergence of an alternating least squares algorithm and properties of limits of calculated sequences are discussed. (SLD)

Develops a testing procedure based on a model including treatment effects on residual variances. (Author/RC)