Accounting for nestedness for ordinal data

[Shu Tian Ng]

Hello all,

I am new to SEM.

I conducted a survey on students at 3 different types of schools, and there are a few schools of each type, giving me a total of 14 sampling sites (I went to all the schools in my sampling frame). The survey comprised scale-type questions, so my raw data is ordinal, and my SEM assesses the relationships between latent constructs after CFA. I want to account for the nestedness of my data but I'm not sure how to do it. I tried using the lavaan.survey package, but got the following error.

Error in t(x) - xbar : non-numeric argument to binary operator

From reading other discussions, I suppose this is because the package cannot work on ordinal/categorical data yet. In that case, is there any way I can account for clustering? Would appreciate it if anyone could point me to a package or some example syntax. 

Shu

[Chao Xu]

Shu,

You are right that lavaan doesn't support multilevel CFA/SEM on ordinal measures yet.

To my knowledge, there is no other R packages that are capable of estimating ordinal multilevel latent variable models. One way to work around ordinal scales is parceling your items. Or you may want to turn to Mplus if you consider multilevel ordinal data analysis is crucial for your project.

Chao