Abstract
This simulation study considered the rule of thumb as noted in Pedhazur (1997) for judging the usefulness of continuous (in a previous MANOVA, dependent) variables at determining group separation in descriptive discriminant analysis; namely, that a structure coefficient value equal to or greater than .3 identifies a useful continuous variable. No research to date has tested this rule. Results indicate that the rule is generally useful for identifying variables with medium to large effects but not small effects.

This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
Copyright (c) 2006 Mercedes K. Schneider (Author)
Downloads
Download data is not yet available.