|
Message/Author |
|
Anonymous posted on Monday, January 30, 2006 - 5:26 am
|
|
|
Hello, I have a quick question. I have a dataset that has a complex design (i.e., weights and clustering). I want to develop a path model at the within-level only. That is, I don't want to include a between-level model, but I want to take my clustering into account. Will Mplus allow me to include my cluster variable in a type=twolevel model without specifying a between-level model? If not what would you suggest? Thanks |
|
|
You can do that. I think you need to include %BETWEEN% in the MODEL command but don't specify a model after it. You could also use TYPE=COMPLEX; |
|
anonymous posted on Monday, January 30, 2006 - 2:06 pm
|
|
|
Thank you |
|
|
Regarding the above q&a, is using the CLUSTER command insufficient for addressing the clustered nature of the data? |
|
|
There are 2 ways to analyze the data with the Cluster command: Type = Complex, which estimates a single-level model, adjusting the SEs to be correct under clustering Type = Twolevel, which estimates a 2-level model (with correct SEs). |
|
Back to top |
|
|