how to set unstructured covariance matrix for random effect in linear mixed models in spss

Question

I would like to ask how to set unstructured covariance matrix for random effect in linear mixed models in SPSS. When I do so in my SPSS (version 26) it is showing me in the output (in model dimension table) that covariance structure for random effects is "Identity" and giving me the bad results compare to the book of Andy Field, where he uses this method. It could be possible, that the versions of SPSS are different as SPSS writes me "As of version 11.5, the syntax rules for the RANDOM subcommand have changed. Your command syntax may yield results that differ from those produced by prior versions. If you are using version 11 syntax, please consult the current syntax reference guide for more information." Thus I am looking for the solution how to set unstructured covariance matrix for random effect in linear mixed models to really get "Unstructured" covariance structure and not "Identity" and thus to get the expected output. Thank you very much for any help.

Answer 1

If you're using the syntax, the COVTYPE specification on the RANDOM subcommand is where you would specify this, using the keyword "UN" to do so. In the dialog boxes, the Random dialog lets you specify this with a drop down menu.

If you're specifying unstructured and the program is changing it to identity on you, it would indicate that the random effect of interest only has one parameter it can fit.

The message about versions 11 and 11.5 is not relevant at this point.