I am currently doing logistic regression.
My independent variables are: Dr. Age, Dr. Race, Dr. Gender, Patient Age,
Patient Race, Patient Gender, length of visit, reason for visit, and
several other variables.
As you can see by running a model with these variables the doctor
variables will be repeated for each patient that doctor has. i.e the
patients are nested within the doctors.
I have ran the model with STATA using the cluster option and I cluster the
Doctor I.D.
for ex: logit exercise drage drrace drgender ptage ptrace ptgender lov
rfv, cluster(did)
what makes this statement different from the following
logit exercise drage drrace drgender ptage ptrace ptgender lov rfv, i(did)
pa robust ?
How also does this effect my interpretation of using the odds ratio for
both methods?
Also, by specifying a cluster it clusters the doctors, but what about the
variables like reason for visit or any patient-related variables, how does
it affect them and their interpretation?
Thanks,
Crystal
*
* 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/
