Assessing the Correspondence of One or More Vectors to a Symmetric Matrix Using Ordinal Regression
- 1 March 1986
- journal article
- Published by Cambridge University Press (CUP) in Psychometrika
- Vol. 51 (1) , 101-112
- https://doi.org/10.1007/bf02294003
Abstract
A statistical model for interpreting psychological scaling research, based on the heuristic work of Reynolds (1983), is developed. This new approach has certain advantages over the standard property fitting approach (Chang and Carroll, 1969) currently used to interpret multidimensional scaling spaces (Shepard, 1962; Torgerson, 1965). These advantages are (a) the ability to directly assess the correspondence of a descriptor vector(s) to a symmetric matrix, and (b) to provide a method in which only ordinal properties of such descriptors are required: thus standard rating, ranking, or sorting data collection methods can be used as the basis to interpret the multidimensional space resulting from the distance data.Keywords
This publication has 9 references indexed in Scilit:
- Efficacies of Measures of Association for Ordinal Contingency TablesJournal of the American Statistical Association, 1978
- A Monte Carlo Investigation of the Statistical Significance of Kruskal's Nonmetric Scaling ProcedurePsychometrika, 1969
- Multidimensional Scaling of SimilarityPsychometrika, 1965
- Multidimensional scaling by optimizing goodness of fit to a nonmetric hypothesisPsychometrika, 1964
- A New Asymmetric Measure of Association for Ordinal VariablesAmerican Sociological Review, 1962
- The Analysis of Proximities: Multidimensional Scaling with an Unknown Distance Function. I.Psychometrika, 1962
- Ordinal Measures of AssociationJournal of the American Statistical Association, 1958
- A Class of Statistics with Asymptotically Normal DistributionThe Annals of Mathematical Statistics, 1948