ABSTRACT
This chapter explores the methods that deal with similarities/dissimilarities data. Similarities data is relational and organized as a square matrix, where the stimuli comprise both the rows and the columns. Similarities data differs from choice data in that it measures how alike/not-alike objects are to each other. Similarities and dissimilarities data are simply reflections of each other. Multidimensional scaling (MDS) methods model these similarities as distances between points in a geometric space. MDS programs are designed to uncover the dimensionality of a given set of data and to visually display the positions of the objects along the latent dimensions. The chapter shows the solution for ratio scale similarities. A ratio scale is an interval scale with a true zero point. The goal of metric MDS procedures is to find a point configuration of inter-point distances that reproduces the observed distances as closely as possible.
