Issue 2
Vol. 43
2001

Multidimensional Scalogram Analysis
of the Gießen-Test (GT)
Martin Grosse Holtforth

Summary
Four questions are examined: Are the old and new scales of the Gießen-Test (GT) or the GT scales constructed by facet theory scalograms? Are differences between items within a scale also qualitative in nature? How can the qualitative variation be described? Which GT items form six-item scalograms? The first three questions are examined by Multidimensional Scalogram Analysis (MSA), the last by cluster analysis on the basis of μ₂-coefficients. No scale was found to be a scalogram on a satisfactory level of error in an MSA analysis of median-split answers of 1541 West-Germans and West-Berliners. Except for one scale all scales had satisfactory two-dimensional solutions. Correlational and cluster analyses could not find other six-item scalograms. The shifting emphasis in the qualitative dimension, possible consequences of these results for the use and revision of the GT, as well as the role of facet theory therein are discussed.

Key words: Multidimensional Scalogram Analysis, Feature Pattern Analysis, Gießen-Test (GT), Facet Theory

Dr. Martin Grosse Holtforth
University of Bern
Department of Psychology
Muesmattstr. 45
3000 Bern 9
Switzerland