Dear Statalisters,
I would like to ask for your advice. I am running a Poisson
regression with glm family(poisson) link(log). When I use options of
pweight and vce (cluster) for the survey setting of sampling weight
and cluster, the output shows quite large Pearson Dispersion, almost
700, which means significant over dispersion. But when I do not use
pweights with the model, the Pearson Dispersion is only 1.25.
The model is like: xi: glm y i.x1, i.x2, x3, …xk [pweight],
family(poisson) link(log) vce (cluster)
My question is why the model with sampling weight produces such a
large dispersion value? Is this simply a sign of extreme over
dispersion or a problem deriving from sampling weight? If this is
because of sampling weight, is there any way to get rid of this large
number?
I will be grateful for your help. Thank you very much for taking time.
Best regards,
Hisako
--
********************************
Hisako Kobayashi
[email protected]
Ph.D Candidate in Public Administration
School of Policy, Planning and Development
University of Southern California
********************************
*
* 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/
