I have prepared my data in SPSS complex sampling. I have applied univariate logistic regression and considered those variables for multivariate logistic regression which have p values less than 0.05 in univariate. But I would like to know, is there any option in SPSS complex sampling where I apply forward selection, backward elimination, or stepwise logistic regression?
Yes there's almost certainly a way to do all of these things but NEVER ever use them because they don't work . Read the attached papers to see why and how they don't work. In addition alternative approaches are mentioned that do work David Booth
The METHOD sub-command of the LOGISTIC REGRESSION command does have FSTEP and BSTEP options. But there are two issues: 1) LOGISTIC REGRESSION is not designed for complex survey data; and 2) old school "stepwise" variable selection methods are ill-advised, as noted by David Eugene Booth. Here are two more resources supporting his statement about that:
- https://www.stata.com/support/faqs/statistics/stepwise-regression-problems/
- https://discourse.datamethods.org/t/author-checklist/3407
However, given that Md. Sahidul Islam is asking about analysis of complex survey data, CSLOGISTIC is the relevant command:
- https://www.ibm.com/docs/en/spss-statistics/28.0.0?topic=reference-cslogistic
From that page:
"CSLOGISTIC performs logistic regression analysis on a binary or multinomial dependent variable using the generalized link function for samples that are drawn by complex sampling methods."
And as the syntax chart shows, there is no sub-command allowing one to use any kind of stepwise variable selection. And that's a good thing, IMO. ;-)
SPSS has many limitations in that sens
you can import your spps file to R
It is more efficient if your are familiar with scripts
As @Bruce Weaver implies some people never get the point. Thanks Bruce for trying anyway. D.
- Badges
- Science topic
- Similar topics
- Psychology