Recent trends in the analysis of categorical or nominal variables were discussed for univariate, multivariate, and psychometric problems. It was shown that several statistical procedures commonly used with these problems have analogues which can be applied to assessing categorical variables. (Author/CTM)

A Monte Carlo study evaluated the effectiveness of three rules of classifying objects into clusters: nearest neighbor classification; nearest centroid assignment; and quadratic discriminant analysis. Results suggest that the nearest neighbor rule is a useful tool for assessing the validity of the clustering procedure of J. H. Ward (1963). (SLD)

Tests for and identifies populations within a family of electronics industries. Data include 669 United States and 144 Japanese electronics firms. Demonstrates the relevance of a general organizational classification for explaining how different natural selection processes affect different populations. (75 references) (MLF)

Predictive discriminant analysis involves a technique used in multivariate classification, i.e., in predicting membership in well-defined groups for units on which multiple measures are available. The validation (assessment) of group membership predictions pertains to two problems: estimating true proportions of correct classifications (i.e., hit…

"Classification: Purposes, Principles, Progress, Prospects" by Robert R. Sokal is reprinted in this document. It summarizes the principles of classification and cluster analysis in a manner which is of specific value to the Marine Corps Office of Manpower Utilization. Following the article is a 184 item bibliography on cluster analysis…

The issues in the interpretation of discriminant analyses presented are restricted to the typical uses of discriminant analysis by behavioral science researchers. Because behavioral researchers use computer programs packages, the issues discussed deal with information obtainable from the package discriminant analysis programs. The following issues…

A methodology developed as an alternative to conventional institutional classification structures, intended to reduce the limitations of those models, is described. Ways in which the methodology can be used for planning, administrative, and research purposes are discussed, as are the dangers in using "peer groups" for institutional…

Numerical classification techniques were used to explore the conjecture that inconsistent results of many studies of disabled readers could result from samples being composed of subgroups of children with different characteristics. Some five subgroups were identified using ITPA scores from a subsample of 37 poor readers. (Author)