CONFIGURAL ANALYSIS,

Abstract

The first section of this paper presents an explication of 'configuration' as it has come to be used in data analysis. In this section a gamut is run--from interaction in the analysis of variance, through factor analysis to general considerations of prediction. The main theme is the nonadditivity or non-linearity of configural information. In section II 'profile analysis' is discussed. There is no particular model associated with profile analysis. It is rather a general approach which has as its goal the separation of groups of similar individuals. Techniques which are frequently used are Q-technique factor analysis and discriminatory analysis. Of some interest is the use of the D measure of Cronbach and Gleser (1953) and 'agreement analysis' proposed by McQuitty (1956). Section III gives a more thorough definition of configural analysis. The configural scale and the polynomial function are given as the two main methods of tapping configural information. Section IV gives the relationship between the configural scale and the polynomial function following Horst (1964). The problem of ascertaining whether configural information is present in a sample of data is briefly discussed and two empirical investigations are cited.

Document Details

Document Type
Technical Report
Publication Date
Sep 01, 1964
Accession Number
AD0607667

Entities

People

  • Gordon E. Wainwright

Organizations

  • University of Washington

Tags

DTIC Thesaurus Topics

  • Agreements
  • Analysis Of Variance
  • Computational Processes
  • Computing-Related Activities
  • Data Analysis
  • Data Mining
  • Data Science
  • Factor Analysis
  • Information Science
  • Linearity
  • Mathematics
  • Polynomials

Readers

  • Business Analytics
  • Regression Analysis.
  • Theoretical Analysis.