At 08:48 AM 1/11/2006, Klugert, Ingmar wrote:
For estimating a certain model I use the procedure oprobit.
Interpreting the results I struggle with the following problem:
I have four country dummies in my model, the p-values say one of
them is somewhat difference to the reference group whereas the
others are not. Taking the cuts given in the output, it now becomes
quite difficult to use the coefficients estimated. Two of the three
countries that have not a different influence on the endgenous
relative to the reference group are lying between cut2 and cut3. The
other one lies, like the country shown as being not similar to the
reference group, below cut2.
Now my question: How is it possible that three countries that are
assumed to be of identical influence on the endogenous by the
estimation results lie in two different groups of cut points.?
Also, the model is not assuming that the groups are of identical
influence. That is your conclusion, based on the significance
tests. If you used, say, the .95 level of significance instead of
the .05 level, you might reach a different conclusion. But as far as
the model is concerned, you are allowing the estimated effects of
each group to differ and hence the range they fall in with regards
to the cut points can differ. If you want to assume that the
influence is identical, then you should drop the dummies for the
groups that you assume do not differ from the reference group.
-------------------------------------------
Richard Williams, Notre Dame Dept of Sociology
OFFICE: (574)631-6668, (574)631-6463
FAX: (574)288-4373
HOME: (574)289-5227
EMAIL: [email protected]
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/
