What is the difference between unstratified and stratidied binary data? How does the logistic model look like for each of them?


Data stratification is dividing into mutually exclusive, collectively exhaustive subgroups. For binary data, the Wikipedia example covers why it can reduce error in estimating the outcome.

For the logistic model, you should apply conditional logistic regression for the stratified case.

  • $\begingroup$ Okay thanks! Which theory are statistical software (such as R and SAS) use when running logistic regression? $\endgroup$ – Greta Apr 13 '17 at 15:42
  • $\begingroup$ It looks like R has a builtin (stat.ethz.ch/R-manual/R-devel/library/survival/html/clogit.html) for conditional logistic regression, you should start there. $\endgroup$ – jbh Apr 14 '17 at 1:06

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.