I'm running a linear mixed model with both factors and covariates, in SPSS. For ease of interpretation of one of the categorical effects, I want to get estimated marginal means for these categorical variables at values of covariates other than the mean. Is there an easy way to do this?
I think the only way to do this is via the SPSS syntax for the procedure MIXED.
To easily do this you could first run your analysis for the model, then copy+paste the syntax from the output window to the syntax window. There you can produce the estimated martinal means with the /EMMEANS subcommand. Specify the TABLES subcommand with the relevant factor and use the WITH subcommand with covariate = value to supply your own covariate level. If you want to compare the factor category estimated means, you can specify a suitable test with the COMPARE subcommand. The full command syntax struture is shown here:
https://www.ibm.com/support/knowledgecenter/en/SSLVMB_23.0.0/spss/advanced/syn_mixed.html
Regards,
Timo
- Badges
- Science topic
- Similar topics
- Philosophy
- Esthetics
- Interpretation