During risk analysis by multiple logistic regression I am facing a problem! Which one to take into account and which is not. Also after risk ratio calculation we can get OR by cross tabulation. So I have to have a clear insight of that!
As I recall, SPSS has some really ugly output for logistic regression, but Helena's tip should sort your odds ratio problem out.
The bigger question – which factors to take into account – is one that you have to solve by thinking.
1. Which are the predictors that are of primary interest to me, the ones that my study hypothesis is about?
2. Which factors might distort the relationships between these predictors and my outcome variable - are there, for example, demographic or clinical differences between the groups I am comparing that need to be adjusted for?
3. Are their other risk factors that are associated with my predictors of interest and with my outcome? What happens when I adjust for those?
That kind of approach…
Hi,
Nice! Plz have time to see these below links-Video have clear instructions how to interpret and the rationale behind the OR could be explore by the second link.
Link to Video- https://www.youtube.com/watch?v=rvopVisrgus
Link to OR- http://www.ats.ucla.edu/stat/spss/output/logistic.htm
Best
OR, RR and Relationship between Or e RR:
Please see the file attached~
In regression models you can only obtain directly OR (=ExpB).
Hello Raja,
The meaning of OR is always the same whatever model you use.
Hi Helena! Thanks for your output! But the file you attached here I am unable to read it! I know only English and Bengali!!
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Statistics
- Statistical Software
- SPSS