I am currently analyzing longitudinal data.
Research question: Do the groups differ with regard to change in proportion with cluster symptom from baseline to post intervention (Time 0 to Time 1), controlling for other important variables?
My question: In GEE (SPSS), which a “working correlation structure" that I need to choose and why?
- AR (1),
- Exchangeable,
- M-dependent,
- Unstructured.
Thank you so much!
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Statistics
- Statistical Software
- SPSS
More Dau Van VU's questions See All
Similar questions and discussions