David M. Levine

A Monte Carlo Study of Kruskal's Variance Based Measure on Stress.

Researchers in the past ten years have studied various parameters involved in nonmetric multidimensional scaling by utilizing Monte Carlo procedures. This paper develops stress distributions using Kruskals second stress formula based upon a null hypothesis of equal likelihood in the ranking of a set of proximities. These distributions can serve to determine whether a set of data has other than random structure.

Nonmetric multidimensional scaling and hierarchical clustering: procedures for the investigation of the perception of sports.

Abstract Nonmetric multidimensional scaling and cluster analysis are described as procedures that can be used to study the perception of sports. Measures of the proximity of objects (sports) are discussed along with alternate data collection methods. Nonmetric multidimensional scaling develops a map of the objects in a varying number of dimensions and enables the researcher to interpret the factors that differentiate the various objects. Hierarchical cluster analysis develops a tree-like structure in subdividing the objects into homogeneous groups. These techniques are illustrated in a study of 13 sports. Two factors that differentiated the sports are revealed by nonmetric multidimensional scaling: team vs individual and degree of action. This result, along with the clusters developed through hierarchical clustering, leads to the conclusion that these sports have been primarily distinguished by how team oriented they were perceived to be.

EFFECTS OF DENSITY OF NOISE FIELD AND DELAY ON INFORMATION-PROCESSING OF PATTERNED INPUTS

The present paper extends current research on pattern recognition to investigate the effects of density of noise field and delay (off) times. The results indicate major effects due to density of the noise field, on time and off time. Recognition initially declined as the density of the interference increased, then subsequently leveled off at higher interference levels. The effect of density of noise field on recognition was not consistent across all levels of either on time or off time. There was a wide range of processing-ability among Ss. Some subjects were capable of high levels of performance for very fast input rates. There was a high degree of consistency among Ss in the relative difficulty of the letters.