M.S. (Master of Science)
Department of Mathematical Sciences
Chi-square test; Contingency tables
This thesis is concerned with comparisons of chi-square tests in an I x J x K contingency table. The hypotheses examined were no-three-factor interaction, conditional independence, and complete independence. Monte Carlo techniques were used to study the behavior of a variety of tests defined by the Crcssie-Rcad family of power divergence statistics for small and moderately sized samples. Results suggest that power divergence statistics in the range λ ε [0.4,1.2] perform well in terms of their nominal values and power. The Pearson Χ² and the Cressie-Rcad 1(2/3) statistic perform equally well. The Χ² has a slight edge over 1(2/3) for each of the three hypotheses.
Cavenee, Patricia L., "A comparison of power divergence statistics for testing hypotheses in three-dimensional contingency tables" (1989). Graduate Research Theses & Dissertations. 178.
xii, 275 pages
Northern Illinois University
Rights Statement 2
NIU theses are protected by copyright. They may be viewed from Huskie Commons for any purpose, but reproduction or distribution in any format is prohibited without the written permission of the authors.