01. SEMinR Lectures Series: Partial Least Squares Structural Equation Modelling (PLS-SEM) in R

Partial least squares structural equation modeling (PLS-SEM) in R is built on a simple foundation: R is the statistical computing language used to import and clean data and then run PLS path models, while RStudio is the integrated development environment that makes writing and executing that code practical. The core takeaway is that getting comfortable with R and RStudio is the prerequisite for doing PLS-SEM work in R—everything from model estimation to mediation, moderation, and higher-order constructs depends on that workflow.

R is described as a free, open-source statistical computing language. Because its underlying code is publicly available, improvements and new features can be contributed by a broad community, which in turn makes R code more reproducible, shareable, testable, and scalable for larger automated applications. The transcript emphasizes that R’s open-source ecosystem includes hundreds of libraries and extensive documentation, so users can find support and add functionality as needed.

In practical terms, R can be run from an operating system command line or directly from the R console. Still, the recommended approach is to use an IDE, since it provides a more efficient coding environment. An IDE is where users write code, manage projects, and run analyses without constantly switching contexts. That IDE in this workflow is RStudio.

The session then turns to setup. Before RStudio can be used, R itself must be installed. After downloading and installing R, users download RStudio—specifically the free “RStudio Desktop” option—and install it as an executable. Once installed, RStudio is opened from the system start menu.

Inside RStudio, the layout is broken into functional areas: a console for running commands, a script editor for writing and organizing code, and panels for items like environment history, connections, and tutorials. The bottom-right area is used for file navigation, viewing plots, and managing packages. Package installation is shown as a concrete example: users can install a package (such as “seminr”) from the install interface, and then see installed packages listed for later use. Documentation and help pages for packages are also accessible from the interface, supporting learning and troubleshooting.

Finally, the transcript frames the broader learning path for the series: after an overview of R and RStudio, the sessions move into structural equation modeling in R using the sem in r package, including evaluation of reflective and formative measurement models, assessment of structural models, and analyses such as mediation, moderation, and high-order constructs. A free, open-access Springer book is recommended as the reference resource for the series, authored by Hair, Thomas Hult, Christian Ringle, Maros A. Sarstedt, Nicholas P. Stanks, and Somia Ray, with Professor Somia Ray noted for answering queries during the series.