University of Illinois Urbana-Champaign

The analysis of free-sorting data: Beyond pairwise cooccurrences

Daws, John Thomas

This item is only available for download by members of the University of Illinois community. Students, faculty, and staff at the U of I may log in with your NetID and password to view the item. If you are trying to access an Illinois-restricted dissertation or thesis, you can request a copy through your library's Inter-Library Loan office or purchase a copy directly from ProQuest.

Permalink

https://hdl.handle.net/2142/20090

Description

Title
The analysis of free-sorting data: Beyond pairwise cooccurrences
Author(s)
Daws, John Thomas
Issue Date
1993
Doctoral Committee Chair(s)
Hubert, Lawrence J.
Department of Study
Psychology
Discipline
Psychology
Degree Granting Institution
University of Illinois at Urbana-Champaign
Degree Name
Ph.D.
Degree Level
Dissertation
Keyword(s)
  • Biology, Biostatistics
  • Statistics
  • Psychology, Psychometrics
Language
eng
Abstract
  • Free-sorting data are obtained when subjects are given a set of objects and are asked to divide them into subsets. Such data are usually reduced by counting, for each pair of objects, how many subjects placed the two into the same subset. The more frequently two objects are sorted together, the more similar the two are inferred to be. These pair cooccurrences, however, do not inform the researcher about the higher-order relationships among the objects.
  • The present study examines the utility of a group of additional statistics, the cooccurrences of sets of three objects. The triple cooccurrence is taken as a measure of how similar the three objects are to each other. Because there are dependencies among the pair and triple cooccurrences, adjusted triple similarity statistics were developed. These statistics are independent of the pair cooccurrences under the null assumption that the subjects are sorting the objects randomly. The characteristics of the adjusted triple statistics under various null and non-null models is explored.
  • Multidimensional scaling and cluster analysis--which usually use pair similarities (or dissimilarities) as their input data--can be modified to operate on triple similarities to create representations of the set of objects. Such methods are applied to two sets of empirical sorting data: Rosenberg and Kim's (1975) fifteen kinship terms and Fitzgerald's (1991) forty rape myths. In both cases, the triple similarities do provide interesting and interpretable information about the subjects' perceptions of the objects they sorted.
Type of Resource
text
Permalink
http://hdl.handle.net/2142/20090
Copyright and License Information
Copyright 1993 Daws, John Thomas

Owning Collections

Graduate Dissertations and Theses at Illinois PRIMARY
Graduate Theses and Dissertations at Illinois
Dissertations and Theses - Psychology
Dissertations and Theses from the Dept. of Psychology