I'm working on a regression analysis which involves eight symptom domains comprised of individual survey items. As is the procedure in my lab, the domain scores are simple averages of the corresponding items.
I started with a stepwise logistic regression procedure to determine which domains best predicted my binary outcome. If a domain was found to be statistically associated with my outcome, I tested the corresponding individual items in a similar stepwise process to determine which items were "driving" the observed associations in the domain regression.
My question is this - are folks aware of any statistical references or research that I could cite which would support this sort of drill-down model building process?
To me it sounds as if an Explanatory Factor Analysis would be appropiate, starting with investigating the association between each domain (8x8 table). But I might be mistaken.
Here is a practical introduction:
https://stats.idre.ucla.edu/spss/seminars/introduction-to-factor-analysis/a-practical-introduction-to-factor-analysis/
I'm afraid that you made some serious mistakes. esp when you used stepwise. See the attached for details. BE SURE TO READ AUSTIN AND TU RIGHT AWAY. Then our paper may give you some ideas on what to do next. Best wishes, David Booth
thank you Jesper Fabricius for the reply! the domains in the instrument we use were actually constructed based on several EFAs (and CFAs) over the last several years.
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Regression