Higher order factor analysis is an extension of factor analysis that is little used, but which offers the potential to model the hierarchical order often seen in natural (including psychological) phenomena more accurately. The process of higher order factor analysis is reviewed briefly, and various interpretive aids, including the Schmid-Leiman…

A desirable property of the equamax criterion for analytic rotation in factor analysis is presented. (Author)

Recent results of Bloomfield and Watson and Knott are used to derive a class of union-intersection tests for sphericity from likelihood ratio tests of independence of two sets of variates. It is shown that the ordinary likelihood ratio test for sphericity has a natural union-intersection interpretation. (Author/RC)

The problem of demonstrating invariance of factor structures across criterion groups is addressed. Procedures are outlined which combine the replication of factor structures across sex-race groups with use of the coefficient of invariance to demonstrate the level of invariance associated with factors identified in a self concept measure.…

A relatively simple technique for assessing the convergence of sets of variables across method domains is presented. The technique, two-step principal components analysis, empirically orthogonalizes each method domain into sets of components, and then analyzes convergence among components across domains. (Author)

Principal components analysis is generalized to the case where any of the variables under consideration can be nominal, ordinal or interval. Hotelling's original formulation is seen to be a special case of this generalization. (JKS)

Provides normative data about the distribution of one measure of similarity of factor loadings, the congruence coefficient, through a Monte Carlo Technique. Matching of "chance" factor patterns was done by the method of Tucker. Statistical tests of the results, based on similarities of the method to canonical and multiple correlation,…

Two proposed methods of factor analyzing a correlation matrix using only the off-diagonal elements are compared. The purpose of these methods is to avoid using the diagonal communality elements which are generally unknown and must be estimated. (Author/JKS)

Factor and cluster analyses are distinctly different multivariate procedures with different goals. However, when used in a complementary fashion, each set of methods can be used to enhance the interpretation of results found in the other set of methods. Simple examples illustrating the joint use of the methods are provided. (Author)

The effects of under and overrotation on common factor loading stability under three levels of common variance and three levels or error are examined. Four representative factor matrices were selected. Results suggested that matrices which account for large amounts of common variance tend to have stable factor loadings. (Author/RL)

Outlined is a model for transformation of one factor matrix to congruence with a second or target matrix in which the correlations among the transformed factors are constrained to certain pre-specified values. Procedures are developed for implementing the model, and are illustrated with example factor solutions. (Author/RC)

The VARAN (variance Analysis) program is an addition to a series of computer programs for multivariate analysis of variance. The development of VARAN exploits the full linear model. Analysis of variance, univariate and multivariate, is the program's main target. Correlation analysis of all types is available with printout in the vernacular of…

An "undesigned" experiment is one in which the predictor variables are correlated, either due to a failure to complete a design or because the investigator was unable to select or control relevant experimental conditions. The traditional method of analyzing this class of experiment--multiple regression analysis based on a least squares…