Thank you Dr. Mehta, your suggestion worked!
Although the model is now running, I am still struggling to figure out how to define my lambda matrices (within-level and cross-level). I have three latent factors at both level 1 and level 2. Each of these latent factors are connected to 18 observed indicators, with 6 indicators presumed to load on factor 1, 6 on factor 2, and 6 on factor 3. I am trying to test a CFA to confirm this factor structure.
When I specify the lambda matrices I do so as follows:
##Level 1 lambda loadings
Lev1_pat <- matrix(c(0,1,1,1,1,1,0,1,1,1,1,1,0,1,1,1,1,1),18,3)
Lev1_pat
Lev1_val <- matrix(c(1,.4,.9,.8,.4,.6, 1,.8,.5,.8,.4,.6, 1,.3,.7,.5,.9,.9),18,3)
Lev1_val
##Cross level lambda loadings
ly12_pat <- matrix(c(0,1,1,1,1,1,0,1,1,1,1,1,0,1,1,1,1,1),18,3)
ly12_pat
ly12_val <- matrix(c(1,.4,.9,.8,.4,.6, 1,.8,.5,.8,.4,.6, 1,.3,.7,.5,.9,.9),18,3)
ly12_val
In each case, the resultant output estimates 48 lambdas when I only want 18 estimates (6 estimates per latent variable). Any advice you have about how I might go about altering my matrix specification would be much appreciated. All of my other matrices appear to be fine (nu, alpha, theta, psi) …once I get the lambdas sorted out, I think it will be ready to go!
Thank you so much for all of your help!