The "k"-means clustering method, while widely embraced in college student typology research, is often misunderstood and misapplied. Many researchers regard "k"-means as a near-universal solution for uncovering homogeneous student groups, believing its success hinges primarily on the selection of an appropriate "k."…

The "p"-median clustering model represents a combinatorial approach to partition data sets into disjoint, nonhierarchical groups. Object classes are constructed around "exemplars", that is, manifest objects in the data set, with the remaining instances assigned to their closest cluster centers. Effective, state-of-the-art implementations of…

Two reduced rank classification procedures, principal components classification and equal weights classification, are described and compared by a simulation study to the standard classification procedure to determine their feasibilities as alternative classification procedures. A justification for the development of these two reduced rank…

A method is described for bifurcating a matrix of objects into two submatrices such that the sum of average deviations of the members from the central tendencies of the two submatrices is minimized. Illustrations of the procedure are included. (DG)

This is a methodological study that suggests a taxometric technique for objective classification of yeasts. It makes use of the minres method of factor analysis and groups strains of yeast according to their factor profiles. The similarities are judged in the higher-dimensional space determined by the factor analysis, but otherwise rely on the…

Study concerned with the effects of procedural differences on the classification of geometrical stimuli. (Author)

The development of multiplicative classificatory skills in children between 4 and 9 years of age was investigated by means of two forms of the matrix-completion task. (Authors)

When performing a classification study, it is often useful to leave a residue of unclassified entities to be analyzed separately. Using an interactional paradigm, theoretical reasoning for this approach is outlined. A procedure--RESIDAN--for conducting a classification analysis using a residue is described, and empirical data are provided. (TJH)

Investigates why children under eight years of age show categorical clustering above chance expectations in free recall, when such organization does not correlate with recall. Six-year-old children and adults were tested for memory of 24 pictures of categorizable items. Proportion of items recalled in category strings and number of strings of each…

Results indicated that availability of categories rather than frequency of words seemed most crucial in determining developmental changes in clustering efficiency. (Authors)

Two experiments explored the differential effects of two kinds of organization (taxonomic and schematic) on retrieval of information. (Author/PN)

This report reviews the personnel literature on the development of job families to illustrate and provide examples of varied approaches to the taxonomic issues of objective, content, and method in job family construction. This booklet first examines the definition of job families and then briefly discusses the potential utility of job families.…

Empirical techniques are developed that may be used in conjunction with data stored in the Institutional Profile System to enhance present capabilities of assessing group structure in medical schools. Relevant literature is reviewed, and the institutionally descriptive data available for analysis and their manipulation into researchable formats…

Eleven indexes of rurality and their effect on the classification of Michigan's 83 counties were investigated. The indexes were based on different concepts of rurality: (1) percent of employment in agriculture, fishing, and forestry; (2) population density and distance to urban centers; and (3) economic conditions. Index rankings were compared to…

The purpose of this paper is to suggest a technique of cluster analysis which is similar in aim to the Interactive Intercolumnar Correlation Analysis (IICA), though different in detail. Two methods are proposed for extracting a single bipolar factor (a "contrast compenent") directly from the initial similarities matrix. The advantages of this…