The issue of whether one should use a single or several random initial configurations in multidimensional scaling is disucssed in this brief note. It is a response to the preceding article (TM 503 491), which commented on another article by Arabie (TM 503 490). (JKS)

Tree representations of similarity data are investigated. Hierarchical clustering is critically examined, and a more general procedure, called the additive tree, is presented. The additive tree representation is then compared to multidimensional scaling. (Author/JKS)

This paper presents a simple random procedure for selecting subsets of stimulus pairs for presentation to subjects. The resulting set of ratings from the group of subjects allows the construction of a group space through the use of an existing computer program. (Author/JKS)

About 92 percent of the children studied had stable, organized cognitive structures for the experimental stimuli while an analysis of the sorting data indicated that only 30 percent of the children had stable structures. (Author/DEP)

The role of conditionality in the INDSCAL and ALSCAL multidimensional scaling procedures is explained. The effects of conditionality on subject weights produced by these procedures is illustrated via a single set of simulated data. Results emphasize the need for caution in interpreting subject weights provided by these techniques. (Author/JKS)

Some methods that analyze three-way arrays of data (including INDSCAL and CANDECOMP/PARAFAC) provide solutions that are not subject to arbitrary rotation. This property is studied in this paper by means of the "triple product" (A, B, C) of three matrices. (Author)

The present paper generalizes the results obtained by Davidson (1972) for nondegenerate stimulus configurations. (Author)

An alternative procedure for converting the multidimensional rank-order data for multidimensional scaling is suggested. Data are converted into triad-comparison data from which the proximities are obtained for multidimensional scaling. (Author/CM)

The kinds of individual differences in perceptions permitted by the weighted euclidean model for multidimensional scaling are more restrictive than those allowed by models developed by Tucker or Carroll. It is shown how problems which occur when using the more general models can be removed. (Author/JKS)

A method based on homogeneity analysis (multiple correspondence analysis or multiple scaling) is proposed to reduce many categorical variables to one variable with "k" categories. The method is a generalization of the sum of squared distances cluster analysis problem to the case of mixed measurement level variables. (SLD)

Comments on the preceding article include "Slicing the Vocational Interest Pie One More Time" (Borgen, Donnay); "Lost in Space" (Harmon); "Alternative Dimensions of the Tracey-Rounds Interest Sphere" (Prediger); "Prestige in Vocational Interests" (Gottfredson); "What Goes around, Comes around"…

Discusses pattern classification of images by computer and describes the Two Domain Method in which expert knowledge is acquired using multidimensional scaling of judgments of dissimilarities and linear mapping. An application of the Two Domain Method that tested its power to discriminate two patterns of human blood leukocyte distribution is…

A model for preferential and triadic choice is derived in terms of weighted sums of central F distribution functions. It is a probabilistic generalization of Coombs' (1964) unfolding model from which special cases can be derived easily. This model for binary choice can be easily related to preference ratio judgments. (SLD)

Multidimensional scaling analysis was conducted on the responses of 648 college students to 110 occupational title items in the Strong Interest Inventory. A three-dimensional structure of vocational interests emerged: Data-Ideas, People-Things, and Sex-Type. (SK)