Hi all, I am reading Andy Field; Tabachnik and Fidell and checking Youtube but am very confused on how to check for this test assumption in SPSS? Is anyone able to share the SPSS steps or the preferred method? Many thanks
In general, you check for linearity or non-linearity in the same way you for for linear regression. For example, you may create a quadratic term (x * x) in include it in addition to x in the (logistic) model. If the quadratic term has a significant effect, this might be a first sign of non-linearity. Or you could convert your x into sets of dummies to check linearity/non-linearity :-)
Hello Rhia,
The most straightforward way (and linearity checking applies only to continuous IVs, by the way) would be to (a) run the logistic model and save the estimated logit scores to your data set as a new variable; then (b) run scatterplots of each continuous IV with the logit scores and judge whether the relationship appears to be linear in form.
Good luck with your work.
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Statistics
- Statistical Software
- SPSS