The purpose of this paper is to examine the capabilities of various procedures for sorting dichotomously-scored items into unidimensional subjects. The procedures include: factor analysis, nonmetric multidimensional scaling, cluster analysis, and latent trait analysis. Both simulated and real data sets of known structure were used to evaluate the…

Argues that co-citation methods combine citing behavior of authors by assuming they share common view of scientific literature which affects assessments of dimensionality and clustering of articles. Co-citation matrices, effects of shared point-of-view assumption, and co-citation compared with bibliographic coupling and centroid scaling are…

Cliff's Index "c" derived from an item dominance matrix is utilized in a clustering approach, termed extracting Reliable Guttman Orders (ERGO), to isolate Guttman-type item hierarchies. A comparison of factor analysis to the ERGO is made on social distance data involving multiple ethnic groups. (Author/BW)

More than one correlation coefficient can be computed between two variables when the data can be organized into subsets determined by one of more grouping factors. Implications of this fact are discussed when the variables are items and the correlations are being computed for purposes of scale development. (Author/

An analysis of self-report delinquency data using four non-metric procedures for structural analysis revealed support for the existence of a general delinquency factor. However, offenses of low seriousness and victimless offenses (drinking and drug-taking items) were only weakly related to this general factor. (Author/CTM)

Describes multidimensional scaling and cluster analysis and the type of problems they can help solve. Cites major literature sources concerning their use. Reviews applications of these methods in rehabilitation research. Discusses possible advantages of nonmetric data-reduction techniques over metric approaches, such as factor analysis. (Author)

Multidimensional scaling (MDS) and its potential use for research and evaluation in health-related professions are discussed. Useful data types, interpretation of results, and various applications of MDS are presented. MDS is less restrictive than factor analysis since it does not assume a linear relationship between the objects/variables of…

Reviews visualization techniques for scientific disciplines and information retrieval and classification. Highlights include historical background of scientometrics, bibliometrics, and citation analysis; map generation; process flow of visualizing knowledge domains; measures and similarity calculations; vector space model; factor analysis;…

One of the major assumptions of latent trait theory is that the items in a test measure a single dimension. This report describes an investigation of procedures for forming a set of items that meet this assumption. Factor analysis, nonmetric multidimensional scaling, cluster analysis and latent trait analysis were applied to simulated and real…

In order to determine the effectiveness of multidimensional scaling (MDS) in recovering the dimensionality of a set of dichotomously-scored items, data were simulated in one, two, and three dimensions for a variety of correlations with the underlying latent trait. Similarity matrices were constructed from these data using three margin-sensitive…

This report contains accounts of studies, about scales to be used in the "A Study of Schooling" research project, undertaken to derive indices for constructs presumed to be measureable by composites of items. The report is introduced by a discussion on the rationale for selecting the research methodology used and an explanation of the…

A procedure for interpreting multiple-discrimination indices from a multidimensional item-response theory analysis is described and demonstrated with responses of 1,635 high school students to a multiple-choice test. The procedure consists of converting discrimination parameter estimates to direction cosines and analyzing the angular distances…

Various methods used to assess the content of a test are reviewed, and a new procedure designed to improve on these methods is presented. The two tests considered are a professional licensure examination, the auditing section of the Uniform Certified Public Accountant Examination, and an educational achievement test, a nationally standardized…