Re: st: NBREG for ordinal scales

[Richard Williams <Richard.A.Williams.5@ND.edu>]

Is the outcome a 5-item ordinal scale, or is it a composite scale 
from ordinal items, i.e. several ordinal variables added 
together?  If the former, why not just use an ordinal method?  Also, 
it might help if you said how the dependent variable was 
worded.  Just from what you have said, it would make me nervous to 
use count methods - it could be that a lot of people have done it wrong!

At 09:29 PM 10/9/2006, Matthew C. Johnson wrote:

> Using the svy program in Stata 9.1, I have a dependent variable that is
> highly skewed, with a large number of zeros. I think negative binomial
> regression is appropriate for my analyses; however, the variable is not a
> count or rate. It is merely a five-item ordinal scale. There are several
> published articles using the same data set, with identical (or nearly
> identical) dependent variables. However, they do not cite a source
> justifying the use of regression techniques for count data when the data
> is a composite scale from ordinal items. Is there a rule on this that I
> have not found in my literature searches? Any help is appreciated
> *
> *   For searches and help try:
> *   http://www.stata.com/support/faqs/res/findit.html
> *   http://www.stata.com/support/statalist/faq
> *   http://www.ats.ucla.edu/stat/stata/
-------------------------------------------
Richard Williams, Notre Dame Dept of Sociology
OFFICE: (574)631-6668, (574)631-6463
FAX:    (574)288-4373
HOME:   (574)289-5227
EMAIL:  Richard.A.Williams.5@ND.Edu
WWW (personal):    http://www.nd.edu/~rwilliam
WWW (department):    http://www.nd.edu/~soc 
*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/