Suggests a way of using factor analytic techniques to supplement multidimensional scaling in such a way as to provide a firm basis for evaluating multidimensional representations. (Author/RC)

Two lines of psychometric interest are combined: a) multidimensional scaling and, b) factor analysis. This is achieved by employing three-mode factor analysis of scalar product matrices, one for each subject. Two of the modes are the group of objects scaled and the third is the sample of subjects. Resulting from this are, an object space, a person…

Several nonmetric multidimensional scaling random ranking studies are discussed in response to the preceding article (TM 503 490). The choice of a starting configuration is discussed and the use of principal component analysis in obtaining such a configuration is recommended over a randomly chosen one. (JKS)

A new method to estimate the parameters of Tucker's three mode principal component model is discussed, and the convergence properties of the alternating least squares algorithm to solve the estimation problem are considered. An example is presented. (Author/JKS)

Multidimensional scaling and factor analysis were applied to Rest's objective test of Kohlbergian moral development. The responses of 160 junior high, senior high, college undergraduate, and graduate students were analyzed in order to investigate whether stage and item scores displayed a hierarchial sequential structure. (Author/SS)

Three methods for grouping items in an opinion survey were compared for their utility in subscale construction: rational organization according to content, factor analysis, and multidimensional scaling. Only subscales based on factor analysis could be refined to meet the criteria of reliability, additivity, and interpretability simultaneously. (SV)

Prior to a three-way component analysis of a three-way data set, it is customary to preprocess the data by centering and/or rescaling them. Harshman and Lundy (1984) considered that three-way data actually consist of a three-way model part, which in fact pertains to ratio scale measurements, as well as additive "offset" terms that turn the ratio…

A maximum likelihood estimation procedure is developed for the simple and the weighted additive models. The data are assumed to be taken by either one of the following methods: (1) categorical ratings--the subject is asked to rate a set of stimuli with respect to an attribute of the stimuli on rating scales with a relatively few observation…

This report describes a test of the robustness of factor-analytic methods in the face of various types of scale transformations on the data. Because of the complexities that would be involved in an exact analytical investigation, the tests were done with simulated sets of data having different factor structures. After factor analyzing the original…

The Cartesian theory of dimensionality (defined in terms of geometric distances between points in the test space) and Leibnitzian theory (defined in terms of order-generative connected, transitive, and asymmetric relations) are contrasted in terms of the difference between a factor analysis and an order analysis of the same data. (Author/CTM)

Two sets of five items each from the Law School Admission Test were analyzed by two methods of factor analysis, and by the Krus-Bart ordering theoretic method of multidimensional scaling. The results indicated a conceptual gap between latent trait theoretic procedures and order theoretic procedures. (Author/CTM)

This study was designed to explain how Profile Analysis via Multidimensional Scaling (PAMS) could be viewed as a structural equations model (SEM). The study replicated the major profiles extracted from PAMS in the context of the latent variables in SEM. Data involved the Basic Theme Scales of the Strong Campbell Interest Inventory (Campbell and…

This paper confronts two interrelated problems, that problem dealing with the psychological concept of attitudes and the problem of attitude measurement. (Author)

A data analysis model called CANDELINC performs a broad range of multidimensional data analyses. The model allows for the incorporation of general linear constraints. Several examples are presented. (JKS)

The computer program WEIGAN makes the weighted G analysis available for computer users. The input and output of the program are described. (Author/JKS)