The purpose of
PLmixed is to extend the capabilities of the
lme4 (Bates, Machler, Bolker, & Walker, 2015) to allow factor structures (i.e., factor loadings, weights, discrimination parameters) to be freely estimated. Thus, factor analysis and item response theory models with multiple hierarchical levels and/or crossed random effects can be estimated using code that requires little more input than that required by
lme4. All of the strengths of
lme4, including the ability to add (possibly random) covariates and an arbitrary number of crossed random effects, are encompassed within
PLmixed. In fact,
optim to estimate the model using nested maximizations. Details of this approach can be found in Jeon and Rabe-Hesketh (2012) doi:10.3102/1076998611417628. A manuscript documenting the use of
PLmixed is currently in preparation.
Citations:Rockwood, N. J. & Jeon, M. (2019). Estimating complex measurement and growth models using the R package PLmixed. Multivariate Behavioral Research. [link]
Jeon, M. & Rockwood, N. J. (2017). PLmixed: Estimate (Generalized) Linear Mixed Models with Factor Structures. R package version 0.1.0. https://CRAN.R-project.org/package=PLmixed.
You can install PLmixed from CRAN in
A vignette containing two example analyses can be found here.