kb/data/en.wikipedia.org/wiki/Facet_theory-7.md

title	chunk	source	category	tags	date_saved	instance
Facet theory	8/8	https://en.wikipedia.org/wiki/Facet_theory	reference	science, encyclopedia	2026-05-05T09:54:10.375224+00:00	kb-cron

== Facet theory: comparisons and comments == Concerned with the entire cycle of multivariate research – concept definition, observational design, and data analysis for concept-structure and measurement, Facet Theory constitutes a novel paradigm for the behavioral sciences. Hence, only limited aspects of it can be compared with specific statistical methods. A distinctive feature of Facet Theory is its explicit concern with the entire
set of variables included in the investigated content-universe, regarding the subset of observed variables as but a sample from which inferences can be made. Hence, clusters of variables, if observed, are of no significance. They are simply unimportant artifacts of the procedure for sampling of the variables. This is in contrast with cluster analysis or factor analysis where recorded clustering patterns determine research results and interpretations. There have been various attempts to describe technical differences between Factor Analysis and Facet Theory. Briefly, it may be said that while Factor Analysis aims to structure the set of variables selected for observation, Facet Theory aims to structure the entire content universe of all variables, observed as well as unobserved, relying on the continuity principle and using regional hypotheses as an inferential procedure. Guttman's SSA, as well as Multidimensional Scaling (MDS) in general, were often described as a procedure for visualizing similarities (e.g., correlations) between analyzed units (e.g., variables) in which the researcher has specific interest. (See, for example, Wikipedia, October 2020: "Multidimensional scaling (MDS) is a means of visualizing the level of similarity of individual cases of a dataset"). Modern Facet Theory, however, concerned with theory construction in the behavioral sciences, assigns SSA/MDS space a different role. Regarding the analyzed units as a sample of statistical units representing all units that pertain to the content-universe, their dispersion in the SSA/MDS space is used to infer the structure of the content universe. Namely, to infer space partitionings that define components of the content-universes and their spatial interrelationships. The inferred structure, if replicated, may suggest a theory in the investigated domain and provide a basis for theory-based measurements. Misgivings and responses One reservation that has been voiced concerns the usefulness of a successful SSA map (one whose partition-pattern matches a content-classification of the mapped variables). What are the consequences of an SSA map? Does such a map qualify as a theory? In response, it may be pointed out that (a) consistently replicated empirical partition-patterns in a domain of research constitute a scientific lawfulness which, as such, are of interest to Science; (b) Often a partition-pattern leads to insights that explain behavior and may have potential applications. For example, the Radex Theory of Intelligence implies that inferential abilities are less differentiated by kinds of material than memory (or rule-recall, see Example 1 above). (c) Faceted SSA is a useful preliminary procedure for performing meaningful non arbitrary measurements by Multiple Scaling (POSAC). See Example 4. A common doubt about SSA was voiced by a sympathetic but mystified user of SSA: "Smallest Space Analysis seems to come up with provocative pictures that an imaginative observer can usually make some sense of –– in fact, I have often referred to SSA as the sociologist's Rorschach test for imagination". Indeed, missing in Facet Theory are statistical significance tests that would indicate the stability of discovered or hypothesized partition patterns across population samples. For example, it is not clear how to compute the probability of obtaining a hypothesized partition pattern, assuming that in fact the variables are randomly dispersed over the SSA map. In response, facet theorists claim that in Facet Theory the stability of research results is established by replications, as is the common practice in the natural sciences. Thus, if the same partition-pattern is observed across many population samples (and if no unexplained counterexamples are recorded), confidence in the research outcome would increase. Moreover, Facet Theory adds a stringent requirement for establishing scientific lawfulness, namely that the hypothesized partition-pattern would hold also across different selections of variables, sampled from the same mapping sentence. Facet Theory is regarded as a promising metatheory for the behavioral sciences by Clyde Coombs, an eminent psychometrician and pioneer of mathematical psychology, who commented: “It is not uncommon for a behavioral theory to be somewhat ambiguous about its domain. The result is that an experiment usually can be performed which will support it and another experiment will disconfirm it. … The problem of how to define the boundaries of a domain, especially in social and behavioral science, is subtle and complex. Guttman’s facet theory (see Shye, 1978) is, I believe, the only substantial attempt to provide a general theory for characterizing domains; in this sense, it is a metatheory. As behavioral science advances so will the need for such theory.”

== References ==

== Further reading == Guttman, R. & Greenbaum, C. W. (1998). "Facet Theory: Its Development and Current Status." European Psychologist, Vol. 3, No. 1, March 1998, pp. 13–36. Levy, S. (Ed.) (1994). Louis Guttman on Theory and Methodology: Selected Writings. Aldershot: Dartmouth. Canter (Ed.) (1985). Facet Theory: Approaches to Social Research. New York: Springer. Guttman, R. (1994). Radex Theory. In Robert J. Sternberg (Ed.), Encyclopedia of Human Intelligence. New York, NY: Macmillan Publishing, 907–912. Hackett, P.M.W. (2021) Facet Theory and the Mapping Sentence: Evolving Philosophy, Use and Declarative Applications, (second, revised and enlarged edition), Basingstoke: Palgrave McMillan Publishers.