I am constructing a multilevel model. To specify, I have two independent variables, one in Level 1 and another one in Level 2. The moderate variable and dependent variable are both in Level 2. Now, the moderate variable will moderate the relationship between independent variable 1 (Level 1), independent variable (level 2), and dependent variable (level 2). I checked some relevant papers and materials indicated this situation should use the MSEM method, the traditional method (MLM) cannot use it to the bottom-up relationship between variables. I just found some code for Mplus to construct a 1-1-2, 1-2-1 model, etc. However, I am not familiar to the Mplus because R is my majority analyzed tool. Hence, I want to ask is it possible for me to use R to construct model? Is there any package that I can use?
Dear Yang Gu,
I've found this package (below). Let's see if it is helpful.
It is pretty old; however, it could help you to find the right track.
Lu, J. G., Page-Gould, E., & Xu, N. R. (2017). Package “MicroMacroMultilevel” Title Micro-Macro Multilevel Modeling. https://cran.r-project.org/web/packages/MicroMacroMultilevel/MicroMacroMultilevel.pdf
refer to the following article. Basically, you can test 1-1-2 as 2-2-2.
you can separate the variance at Level 1 and Level 2, and test Level 2 part for IV and the mediator.
Preacher, K. J., Zyphur, M. J., & Zhang, Z. (2010). A general multilevel SEM framework for assessing multilevel mediation. Psychological Methods, 15(3), 209–233. https://doi.org/ 10.1037/a0020141