I am doing a logistic regression (binary DV) with 13 IV. As a first step I wanted to test the linearity assumption bw. the IVs and the logit. To do this I applied the Box Tidwell Method in SPSS by multiplying ln_independent * IV. Surprisingly all my independt variables turned out to be significant. I think this might be due to my large sample size. My single observations are weighted. Thus, my weighted N~800.000. My unweighted N=10.083. When using the unweighted not all variables are significant but still many of them. How can I solve this problem in SPSS? I have already tried just using a random subsample. However, I am uncertain whether I can just select a random subsample (SPSS Data > Select > Random).

  • $\begingroup$ Better than Box-Tidwell, model all your explanatory variables with splines. How to do that in spss I do not know. $\endgroup$ – kjetil b halvorsen Sep 14 '18 at 12:57

Your Answer

By clicking "Post Your Answer", you acknowledge that you have read our updated terms of service, privacy policy and cookie policy, and that your continued use of the website is subject to these policies.

Browse other questions tagged or ask your own question.