Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: Which output is best for ordered logit?


From   Maarten buis <[email protected]>
To   [email protected]
Subject   Re: st: Which output is best for ordered logit?
Date   Thu, 15 Jan 2009 19:01:39 +0000 (GMT)

--- Johanna Avato <[email protected]> wrote:
> Still, I will also have to come up with output tables, what would you
> use, odds or marginal effects?

You could set up an apperent paradox by presenting both the
continuation odds ratios (which don't change over categories of the
dependent variable), and marginal effects (which change sign over the
categories of the dependent variable) and use the graphs proposed by
Ulrich to solve that paradox. 

This could be a good way of giving an exhaustive interpretation of your
model, but be careful not to put too much emphasis on the paradox as a
key finding in itself; it is a known property of ordered regression
models, see for instance section 5.4 of J. Scott Long (1997) Regression
Models for Categorical and Limited Dependent variables, Thousand Oaks:
Sage.

Hope this helps,
Maarten

-----------------------------------------
Maarten L. Buis
Department of Social Research Methodology
Vrije Universiteit Amsterdam
Boelelaan 1081
1081 HV Amsterdam
The Netherlands

visiting address:
Buitenveldertselaan 3 (Metropolitan), room N515

+31 20 5986715

http://home.fsw.vu.nl/m.buis/
-----------------------------------------


      
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



© Copyright 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index