I apologize if this is a simple question, but when I run my path analysis I have been getting "unknown option:" followed by one of my variables (not a command). When I removed said dichotomous variable, my analysis came back with the error "unknown option:" followed by a different variable (this time a ratio variable). Is this a common issue? I checked, and all my missing cases are -9999 so I don't think that's the issue. Thanks! Kelsey
Thank you, Bengt. That does explain the error, and why I couldn't find anything about the command in the user guide. Is there an alternative for creating lag variables in the present version? I am struggling to find any guidance or examples on this and other forums.
I'm trying to obtain the variance decomposition for a 3-level model but I"m not sure how to get the level 3 variance. I have repeated measures over time nested within patients (Level 2), nested within clinicians (Level 3). I thought mplus automatically accounts for the 2 level nesting of patients over time, so that it is not necessary to include patient as a cluster variable - is that correct?
When I run the syntax for "Type = twolevel random with cluster = clinician ID (i.e the level 3 cluster variable) it only gives me the variance at level 1 (62%) and Level 2 (12%) which i assume corresponds to repeated measures and clients, respectively? How do i get the level 3 variance?
when I run a model with the level 2 cluster variable (in this case client) I get different variances (Level 1 = 54% and Level 2 = 17%). how do i understand the difference in variance in these 2 models.
I then tried running a 3-level intercept only model using this syntax: USEVARIABLES ARE y ; CLUSTER= clinid clieid; TYPE = THREELEVEL RANDOM;
Your Cluster statement has to say that patient is level 2. Perhaps you are thinking of the wide-format approach to repeated measures where you don't have to do that. Also, your %Between% has to use the exact names given in your Cluster statement. If this doesn't help, send your output to Support along with your license number.