An Objective Method of Grouping Profiles by Distance Functions and its Relation to Factor Analysis

Application of an Hierarchical Grouping Procedure to a Problem of Grouping Profiles

“…In discussing their grouping procedure, Sawrey, Keller, and Conger (1960) speak of forming profile groups &dquo;in which the group members are similar to each other and, at the same time, dissimilar from the members of all other groups&dquo; (p. 660). Since their method involves selecting nucleus groups, it appears that, for a specific number of groups, they sought a solution that would maximize the between-group sum of squares and minimize the within-group sum of squares for all profile elements.…”

“…Summary An application of an hierarchical grouping procedure (Ward, 1961) to a problem of grouping profiles is described. The matrix for 25 profiles (based on an art preference test) to which the computer program was applied was taken from Sawrey, Keller, and Conger (1960), who applied a different grouping technique and performed a Q-technique factor analysis of the data. The profile clusters obtained by their grouping technique are compared with those obtained by the hierarchical grouping technique.…”

“…Many different ways of handling the first problem have been discussed in the past decade (Rao, 1952;Cronbach & Gleser, 1953;Helmstadter, 1957; Thorndike, Hagen, et al, 1957;Sawrey, Keller & Conger, 1960). Two approaches to the second problem are currently used (Sawrey, Keller & Conger, 1960). To the best of our knowledge, however, no one has applied an hierarchical grouping procedure to a problem of classifying profiles.…”

A study of standardization of variables in cluster analysis

“…This situation seems at least as likely to occur as the existence of the clusters in the rescaled space. Sawery, Keller, and Conger (1960) were early advocates of the use of the unscaled input data for direct clustering. Second, although the arguments in favor of equal weighting of the input variables may seem appealing, there is no compelling reason to practice democracy while performing all cluster analyses.…”

Optimum Grade Classification with the California Achievement Test Battery

“…The approach is quite generalizable, also, for comparing individuals with one another as well as classifying individuals into groups. Other approaches, such as used by Sawrey, Keller, and Conger (1960), following Cronbach and Gleser (1953), neglect joint distribution probability characteristics (cf., Tiedeman, 1954) and are not recommended by the present authors. The obtained coefficients in Tables 1 and 2 are of the least-squares type and as such are subj ect to change with different subtest variance-covariance matrices.…”