On 17 Aug 2001 10:14:04 -0700, [EMAIL PROTECTED] (JDriscoll) wrote:
> I have a large dataset (N can be 2,000-9,000) with
> mostly categorical outcome variables. Any
> chi square is significant with residuals of 100+
> for tiny differences. I know one can determine
> effect size for continuous variables and show
> result is sign only due to size of the N, but...how
> do I do this for categorical outcome variables?
> Thanks!
"contingency coefficient" is one old description of RxK tables.
For examples that I have had, I am usually interested in
various 2x2 sub-tables. If that suits you, you have a slew of
options: symmetric: e.g., the phi coefficient (Pearson correlation),
kappa, and odds ratio; or dependent on the labeling and roles
of variables: e.g., sensitivity/specificity, etc.
--
Rich Ulrich, [EMAIL PROTECTED]
http://www.pitt.edu/~wpilib/index.html
=================================================================
Instructions for joining and leaving this list and remarks about
the problem of INAPPROPRIATE MESSAGES are available at
http://jse.stat.ncsu.edu/
=================================================================