Hello,
I have a survey design and I wanted to know how I can explain the
variability in the data by logistic regressions, since my response and
predictor variables are all binary, and because the log likelihood and
pseudo R2 are not provided after using svy:logit y x1 x2...
Is there any other way of getting an equivalent for R2 in such a
situation? I am concerned if the Odds Ratios and confidence intervals
in the logistic regressions would have any meaning without an R2 or
its equivalent. The data has been obtained through stratified
one-stage cluster sampling.
Thank you,
Ameya
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/faqs/resources/statalist-faq/
* http://www.ats.ucla.edu/stat/stata/
