Dear colleagues,
we are conducting an observational, longitudinal study with 10 groups that were tested for four times over three years with a metric outcome parameter. The last test’s will be performed in Summer 2021, therefore we do not know everything about the final dataset yet.
The groups differ considerably in size and the drop out rate might be quite high. Thus, it is unlikely that we will be able to use parametric tests for statistical analysis.
Originally, we wanted to use a Mixed-Method-ANOVA to compare how the groups differed over time. Since we believe this is not going to work out, it is our plan to replace the Mixed-Method-ANOVA with a Generalized Linear Mixed Model. However, I find it quite hard to find good tutorials or books on this topic for SPSS. Do you have any suggestions? I would appreciate your help.
Kind regards,
Benedikt Hielscher
Try https://www.pilesofvariance.com/
and Multilevel Modeling of Categorical Outcomes Using IBM SPSS; see
https://tandfbis.s3.amazonaws.com/rt-media/pp/common/sample-chapters/9781848729568.pdf
Try https://www.pilesofvariance.com/
and Multilevel Modeling of Categorical Outcomes Using IBM SPSS; see
https://tandfbis.s3.amazonaws.com/rt-media/pp/common/sample-chapters/9781848729568.pdf
Try Heck et al (2013) Multilevel and Longitudinal Modeling with IBM SPSS. Sometimes working directly with the syntax is easier then with the menu.
I have used MlWin, R, Stata en VARCL to estimate multlevel models, but the menu in SPSS sometimes confuses me.
I'd echo what Peter Moorer said about SPSS and using syntax. However, I really find SPSS cumbersome for these models (though I no longer use it so maybe its improved hugely). Among commercial options Stata is apparently very good, though I mostly use R these days. MlWin is great but I never managed to find an effortless way to get it running on a Mac.
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Statistics
- Statistical Software
- SPSS