Dear Dr. Muthen, I am a PhD student who is interested in multilevel models. I am doing a monte carlo study of a two level model. I need to generate two level-1 variables that have a fixed raw score correlation, say r=.40. But I want to vary the within part and the between part of the overall correlation.
My current plan is to decompose the overall correlation into two matrixes: SIGB and SAMPLE (i.e.,within part). My question is: Can I use Mplus Monte Carlo function to generate data sets based on pre-specified SIGB and SAMPLE correlation matrixes?
I know I can get these two matrixes from raw data, but I do not know if I can go the other way around in Mplus.