CONFIRMATORY INFERENCE AND GEOMETRIC MODELS

Discusses a confirmatory technique that is appropriate for comparing a given geometric model with supplementary data available on the same objects used in the representation. The inference procedure is based on relatively straightforward distribution-free principles and requires the comparison of 1 proximity matrix, possibly reconstructed from a particular geometric model, with a 2nd structure matrix obtained from the supplementary information. Two examples are presented (figural goodness; multidimensional scaling) that illustrate the generality of the same statistical approach. (27 ref) (PsycINFO Database Record (c) 2006 APA, all rights reserved). © 1979 American Psychological Association.