A Monte Carlo study was carried out to investigate the ability of the ALSCAL multidimensional scaling program to recover true structure inherent in simulated proximity data. The results under varying conditions were mixed. Practical implications and suggestions for further research are discussed. (Author/JKS)

An algorithm for assessing the correspondence of one or more attribute rating variables to a symmetric matrix of dissimilarities is presented. It is useful as an alternative to fitting property variables into a multidimensional scaling space. The relation between the matrix and the variables is determined by evaluating pairs of pairs relations.…

A method is presented whereby discrete occupational titles can be represented along a multidimensional continuum reflecting a population's shared perceptions. Their locations on the continuum are used to identify significant others' influence and vocational preferences. The model is shown to be useful for explicating the processes by which…

Reexamined findings of Thompson, Hatchett, and Phillips (1981) which found that females gave more extreme affect ratings and less extreme control ratings to interpersonal verbs than did males. Used multidimensional scaling techniques with 43 female and 39 male college students. Analyses of data failed to replicate findings of sex differences on…

Describes experiment designed to illustrate the potential of multidimensional scaling (MDS) in illustrating the perceptions of those in a group regarding one another, specifically of coalitions. Participants were 35 students enrolled in a graduate business course. Suggests positive results of the exploratory research show MDS holds promise as an…

This study examines demands placed on evaluators by the Joint Committee on Educational Evaluation's standards. Twenty-two experts in evaluation rated the compatibility of implementing pairs of standards or rated individual standards on a series of scales. Multidimensional scaling revealed five demands of ethical practice. (JAZ)

Reports results of experiment to determine whether mental models are constructed of interrelationships and cross-relationships of character attributions drawn in themes of novels and films. The study used "Animal Farm" in print and cartoon forms. Results demonstrated validity of multidimensional scaling for representing both media.…

Presented data that question the ability of both unidimensional masculinity-femininity scales and bidimensional scales to assess comprehensively the multiple factors of sex role identification (N=207). Data supported a multidimensional approach that brings together the reliable factors in both types of scales and provides the possibility of…

A model for the analysis of paired comparison data is presented which is metric, mathematically tractable, and has an exact algebraic solution. (Authors/MB)

This study investigates four widely used methods of collecting direct similarity judgments for nonmetric multidimensional scaling: rating scale, triads, rank order of pairs, and conditional rank order. (Author/PN)

Alternating least squares and optimal scaling are presented as two approaches to the quantitative analysis of qualitative data. A variety of statistical approaches to this problem are discussed. Three examples are presented. (JKS)

The coincidence hypothesis predicts that dissimilarity between objects that differ on two separable dimensions is larger than predicted from their unidimensional differences on the basis of triangle inequality and segmental additivity. The coincidence hypothesis was supported in two-dimensional stimuli studies. (Author/CM)

A family of models for the representation and assessment of individual differences for multivariate data called PINDIS (Procrustean Individual Differences Scaling) is presented. PINDIS sheds new light on the interpretability and applicability of a variety of multidimensional scaling models. (Author/JKS)

A multidimensional scaling analysis of 21 work outcomes yielded a latent structure with three dimensions: value attached by society, level of psychological need met, and extent the outcome is inherent in the work itself. Results suggest a multidimensional model of work outcomes is more useful than a categorical approach. (Author)

A single-step maximum likelihood estimation procedure is developed for multidimensional scaling of dissimilarity data measured on rating scales. The procedure can fit the euclidian distance model to the data under various assumptions about category widths and under two distributional assumptions. Practical uses of the method are demonstrated.…