Re: two nominal variables
- From: Richard Ulrich <Rich.Ulrich@xxxxxxxxxxx>
- Date: Wed, 08 Aug 2007 21:51:23 -0400
On Wed, 08 Aug 2007 05:16:21 EDT, Omar <omarfdawood@xxxxxxxxx> wrote:
Dear Experts,
Good morning
I have two nominal variables, one contain 18 labels and the second with 15 labels, it is a population data not sample, my question is, what is the right statistical technique that can be use?
Do you really have 240 separate questions about
association, or was that simply a convenience of
data collection?
What is your question, and how big is your sample
size?
If these are unordered categories, with a single reply
per subject, you are left with (mainly) a contingency table.
If you want to *test* a few questions that are interesting,
you might want to start by recoding similar categories,
or creating 'miscellaneous' groupings out of tiny categories.
--
Rich Ulrich, wpilib@xxxxxxxx
http://www.pitt.edu/~wpilib/index.html
.
- References:
- two nominal variables
- From: Omar
- two nominal variables
- Prev by Date: Re: Linear regression and Logistic regression
- Next by Date: EM and MCMC: Workhorses for Scientific Computing
- Previous by thread: two nominal variables
- Next by thread: Re: two nominal variables
- Index(es):
Relevant Pages
|
