The Association Graph and the Multigraph for Loglinear Models - Paperback

by Harry J. Khamis (Author)

This practical guide teaches nonstatisticians how to analyze and interpret loglinear models using the multigraph

The Association Graph and the Multigraph for Loglinear Models will help students, particularly those studying the analysis of categorical data, to develop the ability to evaluate and unravel even the most complex loglinear models without heavy calculations or statistical software. This supplemental text reviews loglinear models, explains the association graph, and introduces the multigraph to students who may have little prior experience of graphical techniques, but have some familiarity with categorical variable modeling. The author presents logical step-by-step techniques from the point of view of the practitioner, focusing on how the technique is applied to contingency table data and how the results are interpreted.

Author Biography

Harry Khamis has taught university statistics courses, both undergraduate and graduate, for 30 years. He has directed the Statistical Consulting Center at Wright State University since 1993. Dr. Khamis doctoral thesis involved research on the loglinear model and he authored or co-authored 73 publications, of which 12 deal with the loglinear model. He developed the multigraph representation of hierarchical loglinear models (Journal of Statistical Planning and Inference 53, 63-74, 1996, and related articles); it is now an entry in the Encyclopedia of Statistics in Behavioral Science.