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)

Rotation is used in almost all exploratory factor analysis (EFA) studies. There are numerous rotation strategies that can be employed in these various applications. This paper reviews the various rotation choices in EFA studies, including confirmatory rotation, and presents criteria useful in selecting rotation methods in various analytic…

The concurrent validity of the Achievement Anxiety Test (AAT) and its factor structure are investigated to provide empirical evidence about the quality of AAT. State and trait anxiety were measured by State and Trait Anxiety Inventory (STAI), mathematics anxiety was assessed by a 24-item revised version of the 98-item Mathematics Anxiety Rating…

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,…

To study the applicability of simple structure logic for factorial data from scientific disciplines outside psychology, four correlation matrices from each of six scientific areas were factor analyzed by five factoring methods. Resulting factor matrices were compared on two objective criteria of simple structure before and after rotation.…

A factor rotation scheme is presented which has a property that is interesting for examining factor-structure hypotheses. Given several clusters of tests the orthogonal bounds scheme produces factors which contain these clusters in some optimal sense. (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)

Analyses of interscale relations in 235 college students for the Vocational Preference Inventory and Vocational Interest Inventory, based on interest models of Holland and Roe respectively, suggest that the two-dimensional, circular configuration of occupations of personality types may be an oversimplification which impedes the understanding of…

Rotation of canonical components is contrasted with factor rotation and the concept of communalities in canonical analysis is considered. Three possible strategies for rotating canonical components are described: single rotation of all components, separate rotation of each set, and sequential rotation of one set then the other. Constraints,…

In this paper a rather general theory of oblique factor rotation is outlined. The main results are formulated as four theorems. Necessary and sufficient conditions are derived for two factor matrices to admit identical factor structures and/or factor patterns with factors having unit variances. These conditions are expressed in terms of…

Under some circumstances, it is desirable to compare the factor patterns obtained from different factor analyses. To date, the best method of simultaneously achieving simple structure and maximum similarity is the technique devised by Bloxom (1968). This technique simultaneously rotates different factor patterns to maximum similarity and varimax…

The standardization sample of the Wechsler Adult Intelligence Scale-Revised (WAIS-R) was the basis for an analysis of the WAIS-R subscales in relation to demographic variables. Canonical correlation with orthogonal rotation of composite structural components was performed in order to determine interrelations between both sets of variables.…

The problem considered is that of an investigator sampling two or more correlation matrices and desiring to fit a model where a factor pattern matrix is assumed to be identical across samples and we need to estimate only the factor covariance matrix and the unique variance for each sample. A flexible, least squares solution is worked out and…