A gradient method is used to obtain least squares estimates of parameters in constrained multidimensional scaling in N spaces. Features and constraints of the method and two applications of the procedure are presented. (Author/JKS)

For the ALSCAL multidimensional scaling computer program, it is reported that (1) a new coordinate estimation routine is superior to the original; (2) an oversight in the interval measurement level case has been found and corrected; and (3) a new initial configuration routine is also superior to the original. (Author/JKS)

Values were elicited spontaneously from a sample of undergraduates and adults attending college, and were compared to Rokeach's terminal and instrumental values. Multidimensional scaling revealed a simpler structure among spontaneously mentioned values than Rokeach's values. (JKS)

Nonmetric multidimensional scaling and hierarchical clustering procedures are applied to a confusion matrix of numerals. Two dimensions were interpreted: straight versus curved, and locus of curvature. Four major clusters of numerals were developed. (Author/JKS)

A variety of distributional assumptions for dissimilarity judgments in multidimensional scaling are considered, with the lognormal distribution being favored for most situations. Procedures for maximum likelihood estimation in this setting are described and examples are presented. (Author/JKS)

A new procedure for nonmetric multidimensional scaling is proposed and evaluated in this extensive article. The procedure generalizes to a wide variety of situations and types of data and is robust with respect to measurement error. The statistical development of the procedure and examples of its use are presented. (JKS)

Multivariate models for the triangular and duo-trio methods are described, and theoretical methods are compared to a Monte Carlo simulation. Implications are discussed for a new theory of multidimensional scaling which challenges the traditional assumption that proximity measures and perceptual distances are monotonically related. (Author/GDC)

To avoid reduction in observational data resulting from correlational techniques for analyzing interpersonal interactions, a data-transformation method based on multidimensional scaling techniques was applied to the Family Interaction Coding System. Two resulting dimensions, prosocial-deviance and high-low involvement, were applied to observations…

Under conditions commonly met in optimal scaling problems, arbitrary sets of optimal weights can be obtained by choices of generalized universe scores. It is suggested that the invariant parameters of optimal scaling should be interpreted according to latent trait theory, rather than the arbitrary weights. (Author/JKS)

An experimental procedure involving interaction between subject and computer was used to determine an opitmum subset of stimuli for multidimensional scaling (MDS). A computer program evaluated this procedure compared with MDS based on (a) all pairs of stimuli, and (b) on one-third of the possible pairs. The new method was better. (Author/HG)

A maximum likelihood procedure is developed for multidimensional scaling where similarity or dissimilarity measures are taken by such ranking procedures as the method of conditional rank orders or the method of triadic combinations. An example is given. (Author/JKS)

A method for the least squares regression of one squared variable on a second squared variable when the relationship between the original variables is linear is given. The problem arises in multidimensional scaling algorithms. (Author/JKS)

A strategy for reordering the hierarchical tree structure is presented. While the order of terminal nodes of Johnson's procedure is arbitrary, this procedure will rearrange every triad of nodes under a common least upper node so that the middle node is nonarbitrarily closest to the anchored node. (Author/CTM)

Based on a simple nonparametric procedure for comparing two proximity matrices (matrices which represent the similarities among a set of objects), a measure of concordance (agreement) is introduced that is appropriate when K independent proximity matrices are available. (Author/JKS)