Repository files navigation

An R package for estimating Structural Equation Model (SEM) Trees andForests. They are a fusion of SEM and decision trees, or SEM and randomforests respectively. While SEM is a confirmatory modeling technique,SEM trees and forests allow to explore whether there are predictors thatprovide further information about an initial, theory-based model.Potential use cases are the search for potential predictors that explainindividual differences, finding omitted variables in a model, orexploring measurement invariance over a large set of predictors. Arecent overview is in our latest book chapter in the SEM handbook(Brandmaier & Jacobucci, 2023).

Install the latest stable version from CRAN:

install.packages("semtree")

To install the latest semtree package directly from GitHub, copy thefollowing line into R:

library(devtools)devtools::install_github("brandmaier/semtree")# even better: install with package vignette (extra documentation)devtools::install_github("brandmaier/semtree",force=TRUE, build_opts = c())

Package documentation and use-cases with runnable R code can be found onour github pages:https://brandmaier.github.io/semtree/.

Package vignettes (shipped with the package) contain documentation onhow to use the package. Simply type this in R once you have loaded thepackage:

browseVignettes("semtree")

Theory and method:

• Brandmaier, A. M., & Jacobucci, R. C. (2023). Machine-learningapproaches to structural equation modeling. In R. H. Hoyle (Ed.),Handbook of structural equation modeling (2nd rev. ed.,pp. 722–739). Guilford Press.

• Arnold, M., Voelkle, M.C., and Brandmaier, A.M. (2021). Score-guidedstructural equation model trees.Frontiers in psychology, 11,564403.

• Brandmaier, A. M., Driver, C., & Voelkle, M. C. (2019). Recursivepartitioning in continuous time analysis. In K. van Montfort, J.Oud, & M. C. Voelkle (Eds.), Continuous time modeling in thebehavioral and related sciences. New York: Springer.

• Brandmaier, A. M., Prindle, J. J., McArdle, J. J., &Lindenberger, U. (2016). Theory-guided exploration with structuralequation model forests.Psychological Methods, 21, 566-582.

• Brandmaier, A. M., von Oertzen, T., McArdle, J. J., &Lindenberger, U. (2014). Exploratory data mining with structuralequation model trees. In J. J. McArdle & G. Ritschard (Eds.),Contemporary issues in exploratory data mining in the behavioralsciences (pp. 96-127). New York: Routledge.

• Brandmaier, A. M., von Oertzen, T., McArdle, J. J., &Lindenberger, U. (2013). Structural equation model trees.Psychological Methods, 18, 71-86.

Applied examples (there are many more):

Brandmaier, A. M., Ram, N., Wagner, G. G., & Gerstorf, D. (2017).Terminal decline in well-being: The role of multi-indicatorconstellations of physical health and psychosocial correlates.Developmental Psychology.