[R-sig-ME] New: lmerMultiMember adds support for multiple membership models to lme4

[Jeroen van Paridon]

Hi all,

This is just to inform anyone who might be interested that Phillip Alday and I have developed a small R package that adds support for models with multiple membership random effects to lme4.

Multiple membership models have been discussed on this mailing list several times over the years, but they are not supported in lme4. A few years ago, Ben Bolker very helpfully provided an example of how to �hack� lme4 to insert a multiple membership random effect; we have now reworked that example into a more user-friendly R package that also includes helper functions for setting up multiple membership indicator matrices, visualizing those matrices, and for creating nested multiple membership groupings.

Specifying multiple membership groupings is not difficult, but we recommend people check out the README and vignettes at https://github.com/jvparidon/lmerMultiMember before diving in. Other than that, lmerMultiMember simply extends the functionality of lme4, so model specification, model fitting, and model summaries all work just like in lme4.

This package is not (yet) on CRAN, so it's easiest to install it straight from Github using devtools::install_github("jvparidon/lmerMultiMember")

Feel free to reach out with any questions, notes, or bugs that need fixing!


JP van Paridon, PhD
Postdoctoral Researcher
Department of Psychology
University of Wisconsin-Madison

	[[alternative HTML version deleted]]